COVEY, A. How The Incas Built Their Heartland. State Formation and The Innovation of Imperial Strategies

How the Incas Built Their HeartlandHistory, Languages, and Cultures of the Spanish and Portuguese Worlds This interdisciplinary series promotes scholarship in studies on Iberian cultures and contacts from the premodern and early modern periods. SERIES EDITOR Sabine MacCormack, Theodore M. Hesburgh Professor of Arts and Letters, Departments of Classics and History, University of Notre Dame SERIES BOARD JN. Hillgarth, emeritus, Pontifical Institute of Mediaeval Studies Peggy K. Liss, Independent Scholar David Nirenberg, Johns Hopkins University Adeline Rucquoi, Ecole des Hautes Etudes en Sciences Sociales ‘TITLES IN THE SERIES The Mirror of Spain, 1500-1700: The Formation of a Myth J.N. Hillgarth Bishops, Councils, and Consensus in the Visigothic Kingdom, 589-633 Rachel L. Stocking Toasts with the Inca: Andean Abstraction and Colonial Images on Quero Vessels Thomas B. F. Cummins A Network of Converso Families in Early Modern Toledo: Assimilating a Minority Linda Martz Romans in a New World: Classical Models in Sixteenth-Century Spanish America David A. Lupher The Jesuit and the Incas: The Extraordinary Life of Padre Blas Valera, S.J. Sabine Hyland Upholding Justice: Society, State, and the Penal System of Quito (1650-1750) Tamar Herzog Conflict and Coexistence: Archbishop Rodrigo and the Muslims and Jews of Medieval Spain Lucy K. The Origins of Mexican Catholicism: Nabua Rituals and Christian Sacraments in Sixteenth-Century Mexico Osvaldo F. Pardo Missionary Tropics: The Catholic Frontier in India (16th-r7th centuries) Ines G. Zupanov Jews, Christian Society, and Royal Power in Medieval Barcelona Elka Klein How the Incas Built Their Heartland: State Formation and the Innovation of Imperial Strategies in the Sacred Valley, Peru R. Alan CoveyHow the Incas Built Their Heartland State Formation and the Innovation of Imperial Strategies in the Sacred Valley, Peru R. ALAN COVEY THE UNIVERSITY OF MICHIGAN PRESS Ann ArborCopyright © by the University of Mi All rights reserved Published in the United States of America by The University of Michigan Press Manufactured in the United States of America © Printed on acid-free paper 2009 2008 2007 2006 4 3 2 1 No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, or otherwise, without the written permission of the publisher. A CIP catalog record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Covey, R. Alan, 1974- How the Incas built their heartland : state formation and the innovation of imperial strategies in the Sacred Valley, Peru / R. Alan Covey. p. cm. — (History, languages, and cultures of the Spanish and Portuguese worlds) Includes bibliographical references and index. ISBN-13: 978-0-472-11478-8 (cloth : alk. paper) ISBN-10: 0-472-11478-6 (cloth : alk. paper) 1, Incas—Politics and government. 2. Incas—Kings and rulers. 3. Incas— Antiquities. 4. Political culture—Peru—Cuzco Region. 5. Cuzco Region (Peru)— Politics and government. 6. Cuzco Region (Peru)—Territorial expansion. 7. Cuzco Region (Peru)—Antiquities. I. Title. IL. Series. F3429.3.P65C69 2006 985/.019—deaz 2005026027Preface In the 1550s, Bartolomé de Las Casas wrote a history of the Incas of South America’s Andean region, part of a massive treatise on indigenous cultures of the Americas and their rights as civil societies. Las Casas never visited Peru, and the twenty-seven Inca chapters in his Apologética historia sumaria drew from the reports that travelers and clergy sent back to Spain following the invasion of the Inca empire in 1532. This was arguably the first scholarly treatment of primary sources from the Andes, and in the 450 years since then, hundreds of documents have been written about the Incas, advancing our understanding of the largest native empire to develop in the Americas. Although new colonial-period manuscripts occasionally have been discovered in European or Peruvian archives, some of the most important advances in Inca studies in the past century have come from the field of archaeology, where the physical remains of the Inca empire—its cities, roads, and shrines—have been studied. Inca archaeology has flourished since the 1960s, with a scholarly literature that seems to grow exponentially. Inca scholarship today finds itself at a crossroads, faced with the question of how historians and archaeologists are to collaborate to reconstruct the Inca past. Working as they do with documents written after the fall of the Inca empire, historians tend to project backward in time as far as they feel that they can rely on the historicity of the primary sources—typically to the middle of the fifteenth century, but rarely earlier than AD r4oo. Inca archaeology has traditionally been closely tied to the study of the Spanish chronicles, and much has been learned by collecting data that pertain to the period that the documentary record ad- dresses most explicitly. For archaeology to contribute to Inca studies as an equal to the historical record, however, archaeologists must explain how Inca civilization relates to earlier states and empires in the Andean highlands, and they thus ought to work forward in time from AD 1000 or earlier, Recent investigations by Peruvian, European, and North American archaeologists have begun to explore processes occurring invi PREFACE the Cusco region after the decline of the Wari empire and leading up to Inca imperial expansion (ca. AD 1000-1400). A large body of archaeological data has only recently begun to be employed to bridge the divide between pre-Inca archaeology and historically oriented Inca studies. The principal aim of this book is to unite the historical and archaeological evidence to describe how the Incas became sufficiently powerful to embark on their unprecedented campaigns of territorial expansion, as well as to consider how such developments related to earlier patterns of Andean statecraft. To do so, I propose a new reading of the Spanish chronicles, one that focuses on processes—rather than singular events— occurring throughout the region surrounding Cusco, the Inca capital. The historical record can be used to discuss the kinds of interactions that allowed the Incas to subordinate neighboring groups living in a region of considerable ethnic, political, and economic diversity. The process-based reading of the Inca past can be studied in comparison with a large body of archaeological data, including new data that my colleagues and I collected in 2000 as part of the Sacred Valley Archaeological Project (SVAP). SVAP research consisted of a regional archaeological survey in Peru’s “Sacred Valley of the Incas,” a stretch of the Vilcanota-Urubamba Valley that is famous for its Inca monuments. The regional survey work employed the same field methodology used in two previous archaeological surveys directed by Brian Bauer in the Cusco Basin and areas to its south. Survey crews of three to five Peru- vian and American archaeologists walked in parallel lines of approximately 50 meters through all passable areas of a study region of roughly 400 square kilometers, registering over four hundred archaeological sites on the basis of surface artifacts or architectonic features. Surface collections of pottery were analyzed in the laboratory to identify the locations and sizes of settlements for all phases of the pre-Hispanic occupation of the region. Following the survey research, several important sites were mapped under the direction of Kenneth Sims, and the largest preimperial site, Pukara Pantillijlla, was selected for broad horizontal excavations. Work- ing with crews of twelve to fifteen men from the nearby town of Cuyo Grande, we succeeded in excavating five complete structures, as well as several test units in exterior trash deposits and building interiors. The artifacts and radiocarbon dates from this research permitted us to date Inca expansion into a previously independent area, one that is mentionedPREFACE vii in several Inca histories. The new data also provide a portrait of life at an early Inca administrative center. Archaeological data allow us to identify the physical evidence for regional developments occurring over several centuries, and by reading the Inca histories in a compatible way, it is possible to construct two independent versions of how the Inca heartland was transformed after AD 1000. When synthesized into a single narrative, these two lines of evidence provide us with a detailed account of the rise of Inca civilization. It would be disingenuous to promote a process-based reading of the Inca past without acknowledging that such work would not be possible without the efforts of earlier scholars, as well as a host of people who made the SVAP research possible. I would like to thank the many friends and colleagues who have helped me to develop as an anthropologist and supported me as I conducted the research on which this book is based. My interest in anthropology and archaeology developed while I was an undergraduate at Dartmouth College, and my work with Paul Gold- stein, Deborah Nichols, Roberta Stewart, John Watanabe, and Roger Ulrich introduced me to archaeology, anthropological theory, and the critical study of documentary sources. An initial interest in Maya and Roman archaeology led me to graduate studies at the University of Michigan, but fieldwork opportunities in Bolivia and Peru happily di- verted me to the Cusco region and, eventually, to the SVAP research. At Michigan, I was fortunate to study under a number of distin- guished scholars, many of whom also proved to be generous mentors. Kent Flannery, Jeff Parsons, and Henry Wright taught courses that influenced my thinking about social evolution, the development of political complexity, and processes of state formation. Bruce Mannheim offered a year of Quechua the year before I conducted my fieldwork, which has proven invaluable for studying colonial dictionaries. Sabine Mac- Cormack introduced me to the early colonial documents of non-Andean. South America and has provided much-needed support and research opportunities throughout the dissertation process. Joyce Marcus deserves special mention, acting as my adviser throughout my doctoral studies. Joyce created courses tailored to her students’ intellectual interests and is without rival for her dedication as a mentor and adviser. My doctoral project would not have succeeded without the guidance and criticism of these and other professors.viii PREFACE While Michigan provided the academic and theoretical background for the present study, I owe a debt of gratitude to a number of archaeologists for helping me to become a proficient field archaeologist. I was fortunate to work with Brian Bauer for four field seasons in Bolivia and Peru. While working on his Cusco Valley survey project, I learned survey methodology from Brian, who has provided encouragement, feed- back, and advice as I have ventured out to direct my own fieldwork. Other archaeologists who helped me to develop as a field researcher include Paul Goldstein, Mike Moseley, Chip Stanish, and Adan Umire. Chip has been especially generous with his support and advice and instrumental in helping me to develop manuscripts and grant proposals successfully. The Sacred Valley Archaeological Project received funding from several sources. Predissertation research was supported by the Latin Amer- ican and Caribbean Studies Program at the University of Michigan, as well as a National Science Foundation Graduate Research Fellowship. The fieldwork itself was funded by a Fulbright-Hays Fellowship, a Wenner-Gren Foundation Individual Grant, the Department of Anthro- pology at the University of Michigan, the James Griffin Fund of the Mu- seum of Anthropology at the University of Michigan, and a Discretionary Grant from the Horace H. Rackham Graduate School. Additionally, a National Science Foundation Dissertation Improvement Grant (BCS #0135913) provided funds to process thirty radiocarbon dates collected during field research. The execution of the research would not have been possible without the participation and support of a number of Peruvian colleagues. The staff of the Cusco and Lima offices of the National Institute of Culture (INC) were helpful and patient in reviewing permit proposals and super- vising the field research. The Fulbright office in Lima made it possible to process paperwork and get into the field without major complications, and Cecilia Esparza (the director at the time) made me feel like I had a real support network as I did my research. In the field, I was fortunate to work with an experienced crew of Peruvian archaeologists. Werner Delgado, Ricardo Huayllani, Amelia Pérez, René Pilco, and Herberth Reynaga participated in the survey project, and Ricardo and René supervised excavation crews from Cuyo Grande during the excavations at Pukara Pantillijlla. Dionisio Jancco, the president of Cuyo Grande, helped to organize our excavation crews and was instrumental in the success of the excavations. Wilfredo Yépez served as the Peruvian co-PREFACE ix director of the project, offering sage advice on how to conduct the research. The Yépez family helped out with the project in many ways and has provided friendship and support during my visits in Cusco. Following the fieldwork, I wrote my dissertation at the American Museum of Natural History while working with Craig Morris on data analysis from his work at the Inca city of Hudnuco Pampa. During my time at the museum, Craig, Bob Carneiro, Elsa Redmond, and Chuck Spencer have given encouragement and criticism that have improved my work, I could not ask for a better environment for thinking through the dissertation and completing the writing. Finally, I would like to thank my fellow graduate students for every- thing I have learned from them outside the classroom and in the field. There are too many to name them all, but I would especially like to thank Brad Jones, Stella Nair, Kenny Sims, Tiffiny Tung, Steve Wernke, and Karina Yager for help, support, and good company along the way. In addition to all of these names, I save Christina Elson’s for last. She supported every step leading to the production of this book: the challenges of graduate school, the field research, and the laborious write-up process. Christina offered comments on grant proposals, directed an excavation crew at Pukara Pantillijlla, and read early drafts of text that wasn’t ready for the dissertation committee. Her patience, generosity, and good company helped to make a challenging project possible.Chapter 1. Chapter 2. Chapter 3. Chapter 4. Chapter 5. Chapter 6. Chapter 7. Chapter 8. Chapter 9. Chapter 10. Chapter 11. Contents List of Figures List of Tables A New Orientation for Inca Studies State Formation, Imperial Expansion, and the History of Inca Research in the Cusco Region Ecology and Risk Reduction in the Inca Heartland Wari Imperialism and Local Political Competition in Cusco Setting the Stage for Inca State Formation (AD 1000-1200) The Formation of the Inca State Elite Interaction and Local Responses to Inca Expansion State Expansion, Economic Intensification, and Local Resistance The Imperial Extrapolation of Inca State Expansion Strategies The Transformation of the Inca Heartland Conclusions APPENDIXES Appendix A. Glossary of Spanish and Quechua Terms Used Appendix B. Cusco Regional Chronology Notes References Index xiii xvii 18 37 81 109 138 167 184 208 234 241 246 249 281 321Fig. oe Fig. Fig. Fig. Fig. Fig. Fig. Fig. Fig. ig. 3.3. Fig. ig. 4.3. 2 ede Fig. ig. 4.6. 24.7. 2 4.8. Ter 1.3. 2.1. 22s 23. Dade 25s os Beds 23.4. . 3.50 . 3.6. 367° 2 ede 4.2. 45 + 4.9 - 4.10. Figures Map of Inca territory Pachacutic Inca Yupanqui Modern towns in the Cusco region Important sites and survey projects in the Cusco region The Cusco Basin, ca. 1930 Modern towns in the Sacred Valley study region Sacred Valley sites known from the Spanish chronicles and previous archaeological research The Inca estate at Pisaq Areas of high-elevation grassland (puna) and cordillera in the Inca heartland The Qoricocha puna Upper kichwa plots with beans, tubers, and other cultigens Suni and kichwa lands in the Inca heartland River canalization and terraces at Pisaq ‘Andean peasants sowing potatoes and oca Inca storehouses Estimated territories of the Wari and Tiwanaku polities Qotakalli sites in the Cusco Basin. Formative sites in the Chit’apampa Basin Ragchi, a large ridgetop Formative village Qotakalli sites in the Chit'apampa Basin Qotakalli sites in the Sacred Valley study region Area of dispersed Qotakalli Period settlement Limits of Cusco’s territorial control, AD 400-1000 Pikillacta Araway/Wari Period settlement in the Sacred Valley study region 30 32. 39 4t 42 ae 49 5 57 60 61 62 63 64 65 69 71FIGURES Fig, 4.11. Fig. 4.12. Fig, 5.1. Fig, 5.2. Fig. 5.3. Fig. 5.4. Fig. 5.5. Fig. 5.6. Fig. 5.7. Fig. 5.8. Fig. 5.9. Fig. 5.10. Fig, 5.11. Fig. 5.12. Fig, 5.13. Fig. 6.1. Fig. 6.2. Fig. 6.3. Fig, 6.4. Fig. 6.5. Fig. 6.6. Fig. 6.7. Fig. 6.8. Fig. 6.9. Fig, 6.10. Fig. 7.1. Fig, 7.2. Fig. 7.3. Fig. 7.4. Fig. 7.5. Fig. 7.6. Fig. 7.7. Wanka, a local Wari Period site Wari influence in the Cusco region Chokepukio, located on a low rise across from Pikillacta Killke Period settlement in the Lucre Basin and Urcos area Killke Period settlement in the Cusco Basin Major Killke Period settlements of the Sacred Valley study region Rank-size graphs of the Sacred Valley study region Plan of Markasunay Defensive walls at Muyuch’urqu Map of the location of burnt daub Mortuary structure Mortuary sites in the Sacred Valley study region Aerial view of Minaspata, a Killke Period town in the Lucre Basin Pungurhuaylla, a walled Killke Period site to the south of the Cusco Basin Political complexity in Cusco, AD too0-1200 Allies of Lloque Yupanqui Conquests of Capac Yupanqui and Inca Rocca Known Killke deposits under Cusco Cusco Basin settlement after AD 1000 Settlement hierarchy in the Sacred Valley study region Fortified enclosure at Raqchi The fortified site of Pumamarca Qhapaqkancha, a royal estate The Qallaq’asa complex at Pisaq Administrative building at Pukara Pantillijlla Political and ethnic groups of the Cusco region Distribution of elite titles in the Cusco region Inca marriage alliances The Patabamba area, with Qhapaqkancha in the background Marriage alliances in the story of Yahuar Huaccac Archaeological sites in Huayllacan territory The Inca platform at Qhapaqkancha 76 79 86 87 88 90 92 93 94 95 98 104 107 108 11r 113 123 124 126 128 130 131 132 133 139 144 145 148 149 153 154FIGURES xv Fig. 7.8. Archaeological sites in Cuyo territory 156 Fig. 7.9. The upper Chongo Basin 159 Fig. 7.10. Plan of Pukara Pantillijlla 160 Fig. 7.11. Building 5, a public building at Pukara Pantillijlla. 162 Fig. 7.12. Pisaq, as seen from Pukara Pantillijlla 165 Fig. 8.1. Campaigns of Yahuar Huaccac and Viracocha Inca 168 Fig. 8.2. Caquia Xaquixaguana (today called Juchuy Coscco), the estate of Viracocha Inca 171 Fig. 8.3. Storage at two Sacred Valley sites 175 Fig. 8.4. Inca roads, tampu locations, and gollga remains in the Cusco region 176 Fig. 8.5. Plan of Ancasmarca 178 Fig. 8.6. Structural remains at Ancasmarca 179 Fig. 9.1. Inca imperial expansion 188 Fig. 9.2. Inca expansion in the altiplano 190 Fig. 9.3. Inca expansion in the central highlands 193 Fig. 9.4. Hudnuco Pampa, an Inca administrative city 195 Fig. 9.5. La Centinela, capital of the Chincha polity 199 Fig. 9.6. The pan-Andean pilgrimage center of Pachacamac 200 Fig. 9.7. Inca akllawasi and major shrines 202, Fig. 9.8. Lowland areas, with Inca forts 205 Fig. 10.1. The Citua ritual and the Inca heartland 211 Fig. 10.2. Inca towns invited to an Inca christening 212 Fig. 10.3. Origins of wealth goods in Tawantinsuyu 214 Fig. 10.4. Yana ethnicities in colonial Yucay 216 Fig. 10.5. Inca Period settlement in the Sacred Valley study region 218 Fig. 10.6. Imperial Inca buildings at Caquia Xaquixaguana 220 Fig. 10.7. Modern Calea 222 Fig. 10.8. Agha consumption in Inca ritual 225 Fig. 10.9. Coca lands near Cusco 227 Fig. 10.10. Imperial-period royal estates 229Table 2.1. Table 2.2. Table 3.1. Table 3.2. Table 6.1. Table 7.1. Table 10.1. Table 10.2. Table B.1. Tables Estimates of Integrated State Territory Some Ethnohistoric Indicators of State Formation Processes Ecological Zones in the Cusco Region Wet and Dry Periods in Quelccaya Glacier Cores Inca Conquests Outside of Cusco Architectural Dates from Buildings at Pukara Pantillijlla Colonization of Some Inca-of-Privilege Architectural Dates from Inca Buildings Period Names for the Cusco Region Sequence 24 38 54 114 163 213 221 247CHAPTER ONE A New Orientation for Inca Studies When a small contingent of Spaniards led by Francisco Pizarro and Diego de Almagro invaded the Andean highlands in 1532, the Inca empire was the largest native polity to develop in the Americas. Inca territory included parts of the modern nations of Peru, Bolivia, Chile, Argentina, Ecuador, and Colombia, an area spanning coastal deserts, tropical jun- gles, montane valleys, and alpine tundra (fig. 1.1). That one polity could unite this region and rule a hundred or more distinct ethnic groups and local polities was remarkable to the invading Spaniards. This achievement was made all the more impressive when indigenous informants told them that Inca campaigns of expansion had been made over the span of only a few generations. Many Inca conquests were attested to by eye- witnesses who had fought in Inca armies or were among the conquered. In the decades following the conquest, European interest in the Inca past developed as Spain consolidated its colonial control over the Andean region, and oral histories and eyewitness accounts were collected and ed- ited to describe the Inca past and justify or criticize the Spanish Conquest.' For several hundred years, the historical and administrative texts collected in the first century or so after the conquest provided the sole means for discussing Inca territorial expansion and imperial administration. Many of the most reliable Spanish chroniclers explained Inca imperial origins as the result of the actions of a single dynamic individual named Inca Yupanqui (fig. r-2).2 According to these authors, the Inca capital, Cusco, was invaded by an ethnic group called the Chancas around AD 1438. Viracocha Inca, the Inca ruler, fled Cusco with the town’s popu- lace and his favorite son, leaving behind a young prince named Inca Yu- panqui and a few of his friends and supporters. Before the Chancas could reach Cusco, the creator deity (also called Viracocha) appeared to Inca Yupanqui and promised him victory. Viracocha animated stones on the nearby hillsides, creating a host of warriors who fought in support of theHow the Incas Built Their Heartland » Fig. 1.1. Map of Inca territory, ca, AD 1532. Tawantinsuyu, the Inca realm, was the largest native empire to form in the Americas. Inca prince. The Chanca invaders were repulsed, and Inca Yupanqui then personally crafted a new imperial order. The transformations credited to this ruler—construction of a magnificent urban capital, leadership in ambitious military campaigns, the creation of a legal code, and the establishment of a new religion—were so substantial that he assumed the title Pachacutic (Pachakutiq), “The One Who Brings a New Universal Cycle.”3 Having defeated dangerous enemies and established an imperial order, Pachacutic and his military commanders launched a lightning war upon the Andes that continues to impress modern historians. Although this version of Inca imperial origins makes for compelling reading, it is rather unsatisfactory for explaining how one remarkableELVIOVEMOINGA Nt. redited with innovating 1980 [1615].)4 How the Incas Built Their Heartland and charismatic man could forge a mighty empire out of a village-level society. In trying to separate historical fact from imperial ideology, propaganda, and myth, anthropologists are left asking a number of questions that the Chanca invasion story does not explain adequately. How could a simple group of maize farmers transform themselves suddenly into committed imperialists? Without a centralized government and a large subject population, how could a large army have been assembled and provisioned for long campaigns? If we discount the divine assistance Pachacutic is said to have received from the creator deity, how and why would a group of farmers develop military strategies, a conquest ideology, and administrative policies for subjugating and governing new provinces whose populations outnumbered theirs? If we consider the story of the Chanca invasion carefully, it becomes clear that even the most brilliant and charismatic leader could not have created such a mighty empire single-handedly. Conditions beyond the control of a single individual must have played a significant part. Complicating issues of how the story of imperial origins could be possible is the fact that the Pachacutic account is in part created by se- lective readings of the Inca histories. While certain Spanish chronicles do glorify the actions of Pachacutic above other Inca rulers, several scholars have noted recently that this is because the chroniclers relied on life histories told by certain Inca lineages intent on maximizing their own. prestige during the colonial period. The lineages of Pachacutic’s father, Viracocha Inca, and his son, Topa Inca Yupanqui, also presented versions of the past in which their ancestors were credited with the most important imperial developments.* These rival accounts also fail to explain the processes that made rapid territorial expansion and imperial administration possible. The Inca histories cannot be treated as a complete and factual chronology of events performed by historical individuals. The life stories of individual Inca rulers and their repeated reorganization as state histories led to substantial revisions even before the Spanish Conquest.5 Spanish transcriptions of Inca histories were executed with political ends in mind, and the reconfiguration of Inca narratives to fit European notions of universal history represents a further manipulation of already problematic materials. The comparison of the sequences of rulers in the Inca empire with that in other states and empires reveals the entire king list to be historically suspect, and not just in the preimperial period, as has generally been assumed by many historians.®A NEW ORIENTATION FOR INCA STUDIES 5 Considering that Inca histories are laden with the biases of Spanish and Andean writers, riddled with inconsistencies, and produced to meet specific political and ideological aims, how are modern scholars to dis- cern what really happened? Historians have worked backward in time using the colonial-period documents, leaving off their narratives as they become inextricably intermingled with myth and legend. In this book, I propose an alternate method: rather than attempt to unravel the intri- cacies of Inca historiography, I propose that we cut this Gordian knot by adopting a process-based interpretive stance that treats Inca archaeology and ethnohistory as equals. The colonial historical documents can be used to describe how the Incas interacted with neighboring groups in the generations leading up to the first sustained campaigns of expansion outside the Cusco region. Regional archaeology and excavation data provide an independent perspective on the same period, permitting us to evaluate the historical accounts using the physical remains of actual human activities from that time. A Process-Based Approach to the Study of Inca State Formation This book begins with a comparative look at other early states and empires, as anthropological theory sets the parameters for our interpretive expectations. Rapid territorial expansion like that described for the Incas is not unknown for other empires, but a brief discussion of the Roman and Aztec empires demonstrates that rapid growth is most successful in cases where a centralized state government had already coalesced in the heartland of the emerging empire. Indicators of state formation can be identified in the oral histories and archaeological remains from other known states, and these indicators are used to evaluate the evidence for Inca state formation in the Cusco region.” Studying Inca state formation encourages a process-based reading of the Inca histories. Rather than focus on the singular actions of quasi- historical individuals, this new reading identifies a suite of long-term, qualitative changes occurring throughout the Cusco region over the span of several generations (fig. 1.3). These include the development of a centralized and internally specialized government, the spread of a new religion that reflects and justifies a more hierarchical society, and the emergence of an ideology supporting the permanent conquest of new territories. When looking for evidence for similar processes in the Inca6 How the Incas Built Their Heartland laras @ He INOS AME jects ere * Pe SG Psag” Myint Ee srs SA by SSN Cees i imatambo'. , Gb Si Anta @0 8.6, SO an Ot Sebastian Alp Nl Fig. 1.3. Modern towns in the Cusco region histories, I present information from independent historical sources where possible, privileging those that provide the most detailed discussion of incidents occurring in Cusco and its immediate vicinity prior to the ac- cession of Pachacutic. The Impact of Pre-Inca Civilizations in the Cusco Region While historical studies of the Incas work backward in time using colonial-era documents, this book uses archaeological data to move forward chronologically from around AD 600. At that time, Wari settlersA NEW ORIENTATION FOR INCA STUDIES Zz from the Ayacucho region in the central highlands immigrated to the region and established colonies or enclaves about 35 to 45 kilometers to the southeast of the modern city of Cusco. Wari colonists transformed the area immediately surrounding their settlements over a period of several hundred years, constructing agricultural fields and irrigation canals and investing substantial amounts of labor into building an imperial installation at the site of Pikillacta. The largest Wari sites in the Cusco region were abandoned by AD 1000, at which time any formal political relationship between the Cusco and Ayacucho regions appears to have ended. Excavations at Pikillacta and other Wari sites have led investigators to conclude that the Cusco region was a province of an imperial polity and that the Wari tradition of statecraft and imperial administration must have enabled the rapid rise and expansion of the Inca empire. It is true that Pikillacta was a massive site, one that was probably constructed with the labor of Wari colonists and the local populations of the Cusco region. This would-be provincial capital was never completed or fully occupied, however, making a Cusco province seem more like an unreal- ized goal than an administrative reality. Archaeological surveys of the Cusco Basin and areas to the north and south of Cusco have yielded a very different view of the local effects of the Wari colonial system in the Cusco region. Although the arrival of foreign colonists dramatically transformed the area immediately surrounding new Wari settlements, local groups in other locations continued to live in the same villages and farm the same fields as they had since around AD 4oo. Survey archaeologists have not found evidence of any Wari administrative sites outside of the Lucre Basin and Urcos area. The survey evidence fails to support a provincial model for the Wari occupation of the Cusco region. This is not to say that Wari settlers living near the Cusco Basin were not influential. We know that local people obtained exotic goods like obsidian and turquoise through Wari long-distance exchange networks. Despite the rarity of pottery from the Ayacucho region at sites located more than about ro kilometers from Wari settlements, local groups in the Cusco Basin produced and distributed pottery that emulated Wari design motifs. Even if many local groups were not directly governed by the Wari state, there is evidence that they participated in religious rituals and perhaps public feasts that were sponsored by Wari officials. The Wari polity does not appear to have administered the Cusco8 How the Incas Built Their Heartland region as a province, and the centuries-long hiatus between the abandonment of major Wari installations and the formation of the Inca state makes it unlikely that the Incas simply remobilized a Wari model of statecraft. We must instead look to more indirect influences contribut- ing to the rise of Inca civilization. Wari settlers probably introduced intensive agricultural practices to the Cusco region, as well as social structures whereby labor service was given for public construction projects, then reciprocated through religious rituals and public feasts. The centuries of Wari settlement in the Cusco region must have reconfigured how local groups conceived of the authority of their elite leaders, and such developments set the stage for the political competition that cul- minated in Inca state formation. The Cusco Region after AD 1000 The abandonment of large Wari sites around AD 1000 was part of a major settlement shift throughout the Cusco region at that time. Former occupants of the Wari colonies resettled in a series of towns located near the valley-bottom maize lands that they would have continued to farm. Although systematic survey data have not been collected for this region, it appears that the new towns were probably the principal settlements of small complex polities. The elite leaders of such settlements appear to have organized the same kinds of public works as their Wari predecessors, including defensive walls and religious architecture. The absence of obvious defenses at some towns indicates that these settlements were probably large enough to discourage raiding by hostile groups, but the abandonment of productive farmland in the area between the former Wari colonies and the Cusco Basin has been interpreted as evidence that the balkanized polities that succeeded the Wari colonial system were not necessarily allies of the emerging Inca polity.? The Cusco Basin underwent major settlement changes after AD 1000.!° The population grew dramatically as new groups immigrated into the basin, and the city of Cusco grew over time to become the Inca capital. New irrigation canals and agricultural terraces were built in areas that were previously too marginal for agriculture, and the growing population settled new villages near these fields. Most villages in the Cusco Basin were located on low ridges just above fertile valley-bottom farmland, and these do not appear to have been defended. The overallA NEW ORIENTATION FOR INCA STUDIES 9 settlement system of the Cusco Basin was much more complex and hierarchical than other parts of the Cusco region. Not all sites near Cusco were necessarily under the direct administration of the emerging Inca state; several small fortified villages and ridgetop settlements were occupied in remote areas in the southernmost reaches of the Cusco Basin, most of which appear to have been abandoned before AD 1400. A pattern of high-elevation ridgetop settlement and the construction of defensive works are much more pronounced in the Sacred Valley to the north of Cusco. After AD 1000, settlement became more hierarchical, with a number of small and large villages established at elevations of 4000 meters or higher. Some villages were settled in locations offering natural protection—steep-sided ridges or precipitous cliffs—while others were fortified through the construction of defensive walls and ditches. Where local leadership managed to organize the labor of a community or small polity, defensive projects seem to have taken priority over agricultural intensification or the construction of public or ritual buildings. The largest settlements in the Sacred Valley were located close to a very different set of resources than those in the Cusco Valley, and local populations probably farmed higher-elevation crops (generally, without the benefit of irrigation) and herded camelids. No large settlements were situated near the valley bottom, where some of the world’s most productive maize land is farmed today. Differences in domestic architecture, local pottery decoration, and burial treatment indicate that the Sacred Valley was home to multiple political and ethnic groups at this time. Pop- ulations living on the south side of the valley appear to have interacted frequently with the Incas, while those to the north did not. Groups living to the west of the Cusco Basin also maintained rela- tions with the Incas. Although regional survey work has not been completed in this area, the historical documents describe a number of distinct ethnic and political units. One group, the Ayarmacas, is said to have been a complex polity, with a paramount ruler who owned camelid herds and various agricultural lands and who at some point in the past rivaled the Inca paramount for regional preeminence. The Ayarmacas practiced a mixed herding and maize-farming economy, while several village-level polities are also said to have lived near valley-bottom maize lands located along the important caravan route running west from Cusco to the Apurimac River. It is likely that the villages to the west of the Cusco Basin wereTo How the Incas Built Their Heartland organized similarly to those located to the south of Cusco.!! This area is characterized by low population density, and the inhabitants of the small villages in the region lived near their farmlands in communities that lacked artificial defenses. Groups living between the Cusco Basin and Apurimac River commonly used pottery acquired from the Cusco Basin, and their local settlement and economy remained stable after AD 1000, indicating that their incorporation into the Inca state probably occurred early on and did not result in major transformations of local economies. The entire Cusco region experienced the transformation of settlement patterns, political integration, and local economy after AD tooo. ‘The Inca state thus formed in a region characterized by a high degree of ethnic diversity, on a landscape where a variety of different herding and agricultural strategies were pursued. The maize-farming population of the Cusco Basin grew dramatically, and by AD 1200 or so, there is evidence that the Incas of Cusco were emerging as the most powerful polity in the region—a group that governed a large population and had access to substantiated surpluses of agricultural products with which it could fund state projects. The Incas may have been the dominant group in the Cusco region by the thirteenth century, but they did not rule over most of their neighbors. Many small villages may have lived under Inca protection, and polities of moderate complexity intermarried with the Inca elite and contracted alliances. The strongest rivals to Inca dominance lived to the northwest and southeast of the Cusco Basin. Groups in the latter area were probably descendants of Wari settlers who had their own large settlements and monumental architecture, but they appear to have lacked the large subject population seen in the Cusco Basin. The Formation of the Inca State Given the diversity of the ecological, political, and ethnic landscape after AD 1000, it is not surprising that the Incas employed many strategies to incorporate their neighbors into a single centralized state. Having used the archaeological record to work forward to roughly AD 1200, we can begin to apply our process-based reading of the Inca past for comparison with the archaeological record. The rapid population growth in the Cusco Basin after AD 1000 would have created new administrative needs for the Inca polity, while simul-A NEW ORIENTATION FOR INCA STUDIES 11 taneously presenting opportunities for Inca leaders to increase their power and authority. Labor service provided to the Inca elite was used to build new agricultural lands and irrigation canals, resources that could feed the increasing population but were thought of as the personal estates of Inca rulers and their lineages. The allocation of newly improved farmland would have increased the political power of Inca leaders, undercutting the traditional kin-based authority of local leaders. Some groups may have resisted Inca dominance, but the Cusco Basin ultimately came under the direct control of Inca rulers. As rural populations in the Cusco Basin came to be dominated by Inca paramounts, the growing population living in Cusco would have become increasingly dependent on the Inca elite for organizing an urban food supply. The city grew, flourishing as a center of ceremonial and ritual life for the Cusco Basin, and its sizeable nonfarming population must have grown to include religious specialists, artisans, and officials who served the Inca state. A state religion based on the worship of the sun— to grow the precious maize crop produced on new agricultural terraces— was organized under the auspices of the Inca nobility. As Inca rulers grew more powerful, they began to espouse an ideology justifying the conquest or subordination of neighboring groups. The Inca histories indicate that the first territorial expansion outside the Cusco Basin occurred around the time that the Inca elite consolidated its local control over the basin. It is unclear whether military action was required to incorporate the villages and small polities living to the south of Cusco. Given that people in that area used the Cusco Basin ceramic style and that there is no evidence of major modification of the regional settlement system under Inca rule, it is likely that the incorporation of small groups living near Cusco did not require extended military campaigns. The situation in the Sacred Valley provides a marked contrast. Forts and fortifications were constructed, and as the Incas came to control new territory, they settled new villages in areas between the Cusco Basin and neighboring polities that had been depopulated. Evidence of Inca influence among neighboring groups includes the construction of special architecture and prevalence of Cusco Basin pottery in areas where these were previously lacking. The permanent incorporation of groups living to the north of the Sacred Valley provides evidence that Inca society had developed a central administration with specialized offices for governing groups not sharing kin ties with the Inca elite.es How the Incas Built Their Heartland Inca Expansion Practices and Local Reactions As the Incas began to incorporate the different political and ethnic groups living in the Cusco region, they often resorted to military action, but many groups living to the north and west of the Cusco Basin were initially integrated through more peaceful means. In these cases, marriage alliances prevented major hostilities between groups while at the same time promoting Inca power among these groups. Marriages between Inca rulers and elite women from neighboring groups created last- ing bonds in which the wife’s kin bore the responsibility of providing service to her children and their descendants. This service was reciprocated through Inca sponsorship of festive events, many of which would have taken place in Cusco. As Inca rulers took multiple wives from neighboring ethnic groups, they increased the pool of labor service available for state projects because elite women continued to receive labor service, including the maintenance of lands and herds in their home- towns. The organization of feasts and ritual events to reciprocate labor would have helped to transform Cusco into the center of festive and ceremonial life for the region. Stories of a small polity called the Huayllacans clearly illustrate how the Incas used marriage alliances to extend their dominance. The Huayl- lacans lived just to the north of the Cusco Basin, and archaeological settlement patterns indicate that their population comprised a group of small villages and hamlets whose leaders lived in the large village of Qhapaqkancha. Although elite Huayllacan women had customarily married men from the nearby Ayarmaca group, the Huayllacans broke with this tradition, and an elite woman married the sixth Inca ruler. This provoked a war between the Huayllacans and Ayarmacas, one in which the Incas appear to have remained neutral. Seeing that they were losing to the more powerful Ayarmacas, the local elite permitted the Ayarmacas to kidnap the son of the Inca ruler and his Huayllacan wife. The boy was eventually returned to his father in a sequence of events resulting in additional marriage alliances between the Incas and the Ayarmaca and Anta ethnic groups. Huayllacan treachery was forgiven, but when this group later attempted to meddle in Inca royal succession, the Huayllacans were decimated, and their lands were seized as a personal estate of the Inca ruler whose kidnapping they had arranged when he was a boy. While elite marriage exchange provided a means for extending Inca rule over some of the groups of the Cusco region—particularly thoseA NEW ORIENTATION FOR INCA STUDIES 13 that practiced herding and mixed agriculture in the region to the northwest of Cusco—groups of different ethnicity appear to have been conquered militarily. One of the best-documented examples of conquest and early administration is the case of the Cuyos, a group living to the north of the Sacred Valley. Pre-Inca villages in the Cuyo area were situated at high elevations in the side valleys, an area where mixed farming and herding are likely to have been practiced. Not only does village location imply a different economic base for Cuyo society but this area also has decorated pottery, household architecture, and burial practices that are distinct from that of the Cusco Basin, suggesting real cultural differences. The Cuyos are mentioned as one of the first groups conquered by the Incas, and an Inca governor was placed in the area even though Cuyo elites continued to hold local leadership positions. Archaeologically, the imposition of Inca administration is evident through the construction of special architecture and the prevalence of Cusco Basin pottery at certain villages to the north of the Sacred Valley. In particular, the Incas built religious and administrative architecture at the site of Pukara Pantillijlla, which grew rapidly as it developed into a secondary administrative center of the expanding state. Radiocarbon dates from excavations at this site indicate that the Inca reorganization of the Cuyo area began in the late thirteenth or early fourteenth century. The Cuyos are said to have remained under Inca rule for several generations but were accused of plotting the assassination of the ninth ruler, Pachacutic. In retaliation, Inca armies passed through Cuyo lands, burning villages and killing a large number of the local people. The Cuyo area then became Pachacutic’s personal estate, and valley-bottom lands were developed for intensive maize agriculture and resettled with retainers from conquered provincial regions. The Consolidation of Inca Rule in the Cusco Region If marriage alliances helped to extend Inca dominance and reduce hostilities between different polities, they often failed to produce a peaceful transition to direct Inca rule at the local level. Local rebellion is commonly mentioned as the final phase by which the Inca state consolidated its control over groups in the Cusco region. Because our historical perspective comes from Inca informants, we should be cautious about such accounts. While local leaders would have had good reasons14 How the Incas Built Their Heartland for attempting to throw off the yoke of Inca rule, we must consider the likelihood that the Inca version of events was crafted after the fact to justify the reorganization of local economies and political systems to meet the goals of the Inca elite. To understand why local elites would have rebelled, we should consider the long-term goals of the Inca leadership for the Cusco region. The Inca state had developed a conquest ideology that encouraged expansion, with its related desire to access large agricultural surpluses and exotic goods that were not available locally. Expansion into the Cuyo area provided direct access to caravan routes to the Amazonian lowlands, but it also gave the Inca state control over both sides of a valley where maize could be grown intensively. Sacred Valley groups like the Cuyos and Huayllacans did not farm the valley bottom intensively, so the Incas used labor tribute to canalize the Vilcanota River, construct hillside agricultural terraces, and build irrigation canals that could bring water several kilometers from the side valleys that they now controlled. The transformation of valley-bottom lands for maize production involved massive investments in state infrastructure—not only for the terraces and canals for growing maize but also for storage structures, a road system to bind the area to the capital, and administrative buildings for governing the local population. Inca officials would have de- manded most of the labor service that had previously been available to support local elites. Over time, Inca administrators became the principal authority for deciding how food was produced, loyalty was rewarded, religious rituals were performed, and conflicts were resolved. While cooperation with the Inca state would have been an attractive short-term strategy for local leaders, it undermined the authority and power of these individuals in the long term. From the Inca perspective, local leaders would be increasingly redundant in local administration, and they probably would have represented the principal means for organizing resistance to the complete subordination of local economies, politics, and religious life. Whether the final consolidation of the Cusco region was executed in reaction to local rebellions or was initiated by the Incas themselves as part of a new policy to streamline local administration, it appears that the entire Cusco region was reduced to direct Inca control by around AD 1400. This corresponds to the first territorial conquests made outside the Cusco region—the first campaigns of an emerging empire.A NEW ORIENTATION FOR INCA STUDIES 15 From State to Empire: Strategies for Expansion and Administration By AD 1400, the Incas had been expanding their territory throughout the Cusco region for as much as two hundred years, a period in which they developed strategies for incorporating and governing new populations. Territorial expansion was justified by a sense that the Inca ruler could have no equal, and the Incas often described military action as being an acceptable reaction to hostile responses from local elites. New polities and ethnic groups were initially incorporated with as little use of direct force as possible, and local leaders were commonly left in charge with oversight from Inca officials. If the threat of war was the iron fist for intimidating resistant groups into subordinating themselves to Inca rule, marriage alliances with local leaders and gifts of exotic goods were the velvet glove that encouraged initial cooperation. Once new groups had been brought under state governance, the Incas tended to leave local economies largely untouched, using labor tribute to intensify herding and agricultural production in areas that local communities were not using to their full potential. Facilities had to be constructed for storage, administrative activities, and rituals of the state religion, and newly incorporated territories were linked more closely to the capital through the construction of roads. Some new agricultural lands were allocated for local food production and the propitiation of sacred places and shrines (wak’akuna), but the majority went to support the Inca state, the cult of the sun, and the lineage of the Inca ruler. Over time, state administration superseded many of the functions of local leaders, which may explain why these officials would have led rebellions against Inca rule when opportunities presented themselves. Whether or not the Incas were being truthful in their frequent descriptions of local rebellion, it is clear that over time they preferred to administer territory directly, with Inca officials managing the upper-level political and religious activities, the most important of which took place in Cusco, ‘These patterns of expansion and administration had to be modified as the Incas expanded throughout the Andean region. In the Titicaca Basin, local economies were dominated by herding, and local leaders were wealthy and powerful. On the Pacific coast, several polities already existed where local rulers commanded specialized labor, had a state religion,16 How the Incas Built Their Heartland and were intensively farming most available agricultural land. In Ecua- dor, the Incas encountered groups that had specialized merchants and used shell ornaments in their exchanges. Inca imperial strategies evolved to meet local conditions, and while the maize-farming valleys of the central Andean highlands show strong evidence of direct administration, other regions saw more indirect means of government. It is likely that the Incas would ultimately have desired to implement direct administration throughout their empire, but they clearly lacked the means to do so at the time of the Spanish Conquest. The Transformation of the Imperial Heartland As the Incas conquered farther from the Cusco region, their heartland came to play a different role within the expanding empire. In short, the Inca heartland was promoted above provincial regions as a place apart from the rest of the empire. Most of the autochthonous ethnic groups of the Cusco region were granted honorary Inca status, providing a large population of low-level imperial officials who spoke Quechua, understood Inca state practices, and could be relocated to provincial regions to implement the development of provincial infrastructure and stabilize imperial administration. As local populations were moved throughout the Andes, provincial groups were resettled to the Cusco region as permanent retainers of the Inca elite. The political economy of the Inca heartland was dominated by the royal Inca lineages and the state religion. The descendants of former rulers controlled estates that their ancestors had established, and the retainer population provided a range of services for the Inca elite, including herding, farming, transportation of goods between regions, and the production of cloth. The Cusco region was ethnically diverse, but the retainer population served royal Inca lineages directly. The population of the Inca elite burgeoned in the imperial period, and while some Inca nobles served as high-level imperial administrators, a large number lived on the estates built by their ancestors and meddled in the politics of the imperial heartland. Estates were developed to provide Inca nobles with relaxation and enjoyment but were probably also used to by powerful individuals—especially the wives and mothers of former Inca rulers—to support the influence of competing factions.A NEW ORIENTATION FOR INCA STUDIES 17 Process and the Study of Inca Origins This book proposes that historical or structural readings of the Spanish chronicles alone cannot answer the question of how the Inca empire managed to expand so rapidly throughout the Andean region. Chrono- logical problems in the histories and a “great man” interpretive focus have led to the assertion of social and political underdevelopment in the Inca heartland prior to the rise of Pachacutic. Such assumptions are hard to reconcile with what archaeologists and ethnohistorians in other parts of the world have observed about imperial expansion. The new, process-based interpretation—summarized in the preceding and detailed in the following chapters of this book—can resolve many of the problems with interpreting the Inca past from colonial- period documents. Inspired by the comparative anthropological study of state formation in the archaeological and historical record, this new approach attempts to reconstruct the conditions that made rapid Inca imperial expansion possible. Paradoxically, it is only by turning our at- tention away from the legendary acts of a single Inca ruler that we can understand how the competing interests and deliberate acts of individuals enabled Inca imperialism. By expanding our interpretive vista beyond Pachacutic’s world-changing deeds, we can begin to describe the local effects of Inca expansion, as well as the strategies by which non-Inca elites attempted to maintain local autonomy. A focus on state formation and territorial expansion in the Cusco region permits us to consider how the Inca state came to control large populations and resources, a process through which the strategies used in imperial expansion and administration were developed, tested, and refined.State Formation, Imperial Expansion, and the History of Inca Research in the Cusco Region In promoting a new relationship between historical documents and archaeological data in Inca studies, it is important to state how a focus on regional processes can be preferable to the research orientations that have been influential in recent decades. The Inca empire is one of the best-known pre-Hispanic cultures of the Americas thanks to the research of anthropologists and historians, and a new interpretive paradigm would not be possible without such research. The process-oriented approach advocated here is not intended to negate or replace the existing scholarly literature on historiography or the structural characteristics of imperial Inca society. Where an interdisciplinary study of long-term regional processes can prove useful is in providing a narrative of the kinds of changes that are most likely to have occurred in the Cusco region, developments that enabled the trajectory of rapid territorial growth throughout the Andean region after AD 1400. While the interpretations presented in this book are based on sixteenth-century historical accounts and archaeological data, its expectations and interpretation are driven by propositions derived from the comparative study of other ancient states and empires. This chapter describes the comparative anthropological approach pursued in this book, presenting a suite of archaeological and historical indicators that guide the interpretive process. Once the theoretical expectations have been articulated, a brief overview of Inca studies in the imperial heartland demonstrates the intellectual debt that this book owes to previous researchers, as well as the ways that it diverges in its conclusions. 18STATE FORMATION, IMPERIAL EXPANSION 19 Anthropological Perspectives on Imperial Expansion Many authors remark at the rapidity of Inca imperial expansion, as though it stands out distinctly from that of other early empires.! In fact, comparisons with other empires indicate that rapid territorial expansion is a common characteristic of empire building but that the conditions making such expansion possible require an anthropological explana- tion. While the official histories of imperial societies often focus on the agency of a single individual in establishing an empire—an Alexander the Great, Genghis Khan, or Pachacutic Inca Yupanqui—anthropologists recognize that to achieve monumental results, the actions of even the most brilliant and charismatic individuals are dependent on prevailing ecological and social conditions.? In short, it is more significant to understand the development of particular conditions favoring social transformation than to develop biographies of the charismatic individuals who come to be credited with such changes. Rather than debate the historical role of Pachacutic or any other Inca emperor, we should attempt to understand how an Inca polity developed that could take advantage of particular political and ecological conditions in the Andes to expand so rapidly. A cursory look at a few other empires that achieved rapid territorial expansion provides some important clues for reconstructing the conditions under which the Inca empire developed. Early Roman Imperial Expansion (ca. 264-134 BC) The first wave of Roman imperial expansion followed several hundred years of warfare and alliance building between Rome and its Etruscan, Latin, and Samnite neighbors (among others).+ During the early third century BC, Rome defeated rebellions by Etruscan towns, repulsed Gal- lic invasions, and fought a prolonged war with the Samnites, becoming mired in the politics of southern Italy, which had been dominated by Greek colonies for several hundred years.* As Rome continued to consolidate its control over central Italy, Roman armies were sent to southern Italy in support of allies and were garrisoned in several towns. A fear of growing Roman power led some Greeks living in southern Italy to seek the aid of Pyrrhus, the king of Epirus, who invaded the Italian peninsula in 280 BC. Pyrrhus marched to within 60 kilometers of Rome with a professional army of twenty-five thousand men—and twenty elephants—and the support of the Greek cities of Tarentum, Metapontum,20 How the Incas Built Their Heartland and Heraclea, as well as the Samnites, Lucanians, Bruttians, and Mess- apians. Most Roman allies failed to defect to the Greek cause, however, and after several costly victories over Roman armies, Pyrrhus abandoned Italy, leaving Rome the undisputed power on the peninsula. The first war with the Carthaginian empire began just a decade later. Erich Gruen provides a brief description of Rome’s position in 265 BC, on the eve of its first major territorial expansion outside of the Italian peninsula: The Italian peninsula was now united under Roman hegemony. Through absorption, annexation, common citizenship, or treaty the Italian states were bound to Rome in common policy. . .. The relative isolation of Italy from the areas conquered by Alexander gave Rome the time needed to extend her influence and consolidate her holdings at home. The city’s central location in the peninsula, astride the Tiber, afforded a unique position for dominion and control. And Italy’s central position in the Mediterranean basin enabled Rome to face both east and west. By 265 Roman legions, drawn from the peasantry and fighting for their city and homes, had already proved their skill under fire for generations. Rome won the loyalty of Italy, not only through force of arms, but through extension of the franchise and sharing of institutions.” From 264 to 134 BC, Rome expanded rapidly, conquering a rival empire, Carthage, and incorporating its territories as new provinces. It also annexed the city-states of Greece and Asia Minor into this developing imperial order. In roughly the same span of time that Inca territory expanded to control much of the Andean region, Rome rose from a leading power in the Italian peninsula to become the preeminent power in the Mediterranean world.* The critical invasion of Pyrrhus (280-75 BC) may have been important for determining the tempo by which the final resistance to Roman order in Italy was defeated, but like the Chanca invasion of the Cusco region—said to have been the stimulus for Pacha- cutic’s transformation of the Inca polity—this appears to have occurred toward the end of a long regional integration process.’ The Aztec Triple Alliance (c. AD 1428-1520) ‘The period leading up the rise of Aztec civilization remains shrouded in myth and imperial propaganda." Like the Incas, the Mexica elite re-STATE FORMATION, IMPERIAL EXPANSION 21 counted what Michael E. Smith describes as a “rags-to-riches story,” a myth involving migration from a distant origin place by a small group that was destined to rise to dominance over the many Postclassic polities of central Mexico.'! Through alliance manipulation and the use of force, the Mexica and their allies consolidated their control over most of the complex polities (altepet!) in the Basin of Mexico between about AD 1250 and 1430.!2 Stories of Mexica origins describe the initial settlement of their capital, Tenochtitlan, which grew rapidly and came to be supported by intensive raised-field agriculture. Although the Mexica were originally subordinates to the nearby Tepanec polity, they eventually appointed as ruler (tlatoani) a man said to be descended from the Toltecs. Smith describes the rule of Huitzilhuitl(r. AD 1391-1415) in the final decades leading up to the emergence of the Aztec Triple Alliance: Under his popular leadership, people came from all over the Valley of Mexico to live in Tenochtitlan, and the city expanded greatly. The Mexica became highly skilled as soldiers and diplomats in their deal- ings with neighbors. One of Huitzilhuitl’s major accomplishments was the establishment of successful marriage alliances with a number of powerful dynasties. . . . In addition to his diplomatic success, Huitzilhuitl also led the Mexica to victory in a number of military campaigns.!3 In 1428, when the Mexica under Itzcoatl successfully rebelled against Tepanec domination, their polity was already wealthy and powerful, with a reputation as able diplomats and fierce warriors. The Mexica and their allies quickly incorporated the complex polities of the Valley of Mexico, then extended their territory rapidly for roughly ninety years, incorporating large parts of Mesoamerica into their provincial system.'* When Hernéan Cortés reached the Gulf Coast in 1519, the Aztec empire was the largest polity to develop in Mesoamerica. Prior to its short phase of imperial expansion, Mexica society developed several important features, including an urban capital, a system of intensive agriculture that could feed a large, nucleated population, and a government-supervised marketplace for acquiring and exchanging staples and exotic goods. The Mexica elite held specialized positions, achieving status as warriors, diplomats, and merchants. Marriages were the basis for many alliances with groups in the Valley of Mexico, as well as polities beyond that region.» 8 How the Incas Built Their Heartland Conditions Enabling Rapid Imperial Expansion We could discuss many other examples of rapid imperial expansion, but the preceding cases are sufficient to demonstrate not only that other early empires managed to incorporate vast new territories in 100-150 years but that the polities embarking on such campaigns had developed a certain degree of hierarchy and complexity.!* Many anthropologists would identify the kinds of changes described for the preimperial Ro- man and Mexica polities as processes resulting in the formation of a centralized state. State Formation and Its Archaeological Indicators Anthropological archaeologists have debated the nature of state formation and have presented several means for evaluating political complexity in the archaeological record.'® For this book, I define a state as a polity characterized by a centralized and internally specialized government.” Archaeological indicators of state formation can be identified at several different interpretive levels, and such identifications are most secure when multiple indicators are present.'® State formation often takes place within a relatively small region and is often followed by territorial expansion (table 2.1). At the regional level, the development of a four-tier settlement hierarchy and a log-normal or primate rank-size graph have been interpreted as evidence for increasingly hierarchical and centralized organization,'? an apparatus settlement system considered to be consistent with an administrative hierarchy that is more complex than those TABLE 2.1. Estimates of Integrated State Territory Name Integrated Territory Dates Source Susa Up to 20,000 km? Middle Uruk Wright 1998 Warka 15,000+ km? Early Uruk Johnson 1980, 248 Monte Albin — 20,000 km? MAIL Marcus and Flannery 1996 Tikal 20,000+ km? Early Classic Adams and Jones 1981 Teotihuacin —_25,000-50,000 km* Early Classic Marcus 1998 Rome 20,000+ km? Fourth to third Harris 1979 century BC Maradi 8,000 km? 1854-75 Smith 1967 Dahomey 24,000 km? Nineteenth century Spencer 1990 Benin About 10,000 km? Nineteenth century Bradbury 1967STATE FORMATION, IMPERIAL EXPANSION 23 of prestate societies. Log-normal rank-size graphs are often taken as evidence of centrally coordinated regional economies, but in cases where the urban capital of the state is disproportionately large, the graph may be convex.?° Ac the site level, the construction of palaces and appearance of standardized religious and administrative building forms can indicate the establishment of state institutions.?' The leaders of prestate polities are typically not capable of using labor tribute to construct palatial residences, and the standardization of religious and administrative architecture is thought to accompany the development of religious and administrative hierarchies. Qualitative changes in the presence of exotic materials and craft products might also be considered evidence of economic centralization or social stratification concomitant with state formation. The archaeological evidence for other cases of state formation suggests that multiple indicators are needed to demonstrate such processes unambiguously.22 The success of the archaeological model depends on the existence of sufficient data to identify multiple indicators of Inca state formation. Oral Histories and Processes of State Formation In chapter 1, several problems with the historicity of the Spanish chronicles were noted. Narratives of the past were reorganized repeatedly in Inca times, and their transcription decades after the conquest was often influenced by an author’s political intentions, philosophical convictions, or notions of proper historiography. The colonial period documents describing the Inca past cannot be treated as a complete and factual chronology of events performed by historical individuals, and scholars ought to explore other useful ways of studying them. The archaeological record provides one means for evaluating what kind of society the Incas lived in at the time of their initial imperial expansion, and it is possible to read the historical documents in a compatible way to identify the same processes of state formation described in the archaeological record. Just as the archaeological indicators of state formation enumerated in the preceding are derived from the comparative study of several archaeologically known states, it is possible to identify indicators of state formation in the oral histories of societies known to have experienced such processes. Using the same basic definition as for archaeological24 How the Incas Built Their Heartland cases—a suite of qualitative changes involving the emergence of a centralized, internally specialized government—it is possible to identify state formation for several societies, including Dahomey in West Africa, the Betsileo and Merina of Madagascar, and Hunza in Kashmir. Some of the kinds of innovations mentioned in the oral histories of these cultures are presented in table 2.2, which lists the numbers of rulers credited with a particular development. Kent Flannery has recently observed that the role of human agency becomes more pronounced when we consider historical cases of state formation.23 In looking at long-term, regional processes, we do not focus on the “state makers” glorified in the oral histories but instead identify how a given polity changed internally and in relation to its neighbors. From the table, it is possible to see that emerging states often develop more intensive economies in which production is centrally managed through laws and religious ceremonies that are controlled by the elite. Some Ethnohistoric Indicators of State Formation Processes Merina _Betsileo Dahomey Hunza Territorial control of over 2000 km? Ruler; Ruler7 Rulerg Ruler 4 Reduction of local autonomy 3 47 4 4 Growth of capital immigration) 3 47 2 4 Permanent military conquests 3 5 2 2 Codification of judicial system NA 4-5 3 4 Imposition of tribute requirements NVA ss 3 4 Complex elite hierarchies with more titles 2 5 4 4 Asymmetrical marriage alliances 1 NIA 1 NIA Construction of palaces or royal estates 3 NIA 3 NIA Intensification of production 3 s7 N/A 4 Sumptuary laws, direct state access to exotics 2 5 3 4 Calendrical reform, management of specialized production 1 NIA NIA 4 Co-option of local religion or rituals 23 NIA 3 3 Development of state religion and religious hierarchy 23 5 3 3 Introduction of technological innovations NIA N/A 4 Introduction of military innovations m2 5 3 NIA Note: Numbers indicate which ruler (or rulers) in a multigenerational sequence is credited with a given development. For the Merina, ruler r is Andriamanelo (see Brown 1979, 120-30 for sequence). Kottak (1980, 67-81) discusses Betsileo state formation, noting that the process began with rulers 3 and 4, a woman and her son, Herskovits (1938, 11, 13) discusses several ruler lists for Dahomey; for this table, ruler 1 is Dogbagri, and the account of Argyle (1866, 313) is used. The Hunza case uses information in Sidkey (1996) and considers Mir Silim Khan as ruler 4 in the process of state formation.STATE FORMATION, IMPERIAL EXPANSION 25 Immigration to the capital increases, and increased agricultural production feeds a growing urban population of retainers, craft specialists, and bureaucrats. The management of a centralized economy and large population requires special administrative and religious offices, and elite hierarchies become more complex as the state forms. New intermediate elites are supported by tribute, tax collection, or labor service, which is also used by the ruling elite to build palaces and private estates, as well as to acquire exotic goods not available locally. Externally, the emerging state begins to conquer or annex new territory, reducing previously autonomous neighboring elites to subordinates. Conquests are often made in order to acquire productive resources or control over important trade routes. The permanent incorporation of new territory requires that new provincial officials be appointed. As with archaeological indicators for state formation, it is the identification of a whole set of changes—rather than the innovation of any single indicator—that will help us to recognize the development of a centralized state government. We are seeking evidence of a fundamental change in the nature of Inca society, as well as the ways that the Incas interacted with neighboring groups. Since multiple versions of the Inca past exist, we should be able to identify the same kinds of changes in independent sources.24 The most useful sources for this analytical process will be those that provide detailed information on events occurring within about 50 kilometers of Cusco prior to the reign of Pachacutic. While the archaeological evidence provides the means for evaluating the historical sources, it is critical to avoid selecting only the information that corroborates our anthropological expectations. Process-Based Perspectives and the Study of the Inca Past Although the formation of an Inca state exhibiting the same traits as the cultures in the comparative study would have major significance for anthropologists focusing on long-term changes in regional databases, it would have had far less relevance to the sixteenth-century Incas and Europeans whose interest in the Inca past was oriented toward contest- ing the time depth of the imperial period. Many writers attempting to diminish the Inca imperial achievement recount incidents relevant to the study of state formation. The repeated reorganization of Inca histories26 How the Incas Built Their Heartland certainly would have affected how the processes of state formation were remembered, but this becomes less important if multiple versions of the Inca past describe the same kinds of changes observed in other cases of state formation. Even though the interpretive approach proposed in this book takes a methodological departure from most modern historical studies, there is a long tradition of process-based studies of the Incas. Polo de On- degardo, one of the most astute early colonial administrators, observed that “these Incas conquered through violence and warfare . . . and thus the only difficulty was in conquering those neighboring polities in the Cusco region, because afterwards, all the conquered groups went with them [the Incas], and it was always a much larger force than the others [i.e., opposing local groups].”25 In 1571, Ondegardo speculated that Inca imperialism had developed through processes occurring over 350-400 years in which the region within about 25 kilometers of Cusco was consolidated under Inca control.?° Pedro Sarmiento de Gamboa concluded that Inca strategies for expansion and consolidation were transformed under Pachacutic’s father following several generations of raiding and indirect control over the strongest rivals to the Inca polity.?” If some colonial writers—especially those who developed synthetic prespectives on the Inca past using multiple lines of evidence—recognized that local processes in the Cusco region had somehow made Inca imperial expansion possible, why have modern scholars fixed upon the story of Pachacutic and the Chanca invasion as the most reasonable explana- tion of imperial origins? The simplest answer to this question is that the Pachacutic story provides the most reasonable paradigm for describing Inca imperial expansion, regardless of its shortcomings in addressing the processes leading up to such expansion. Although many of the most reliable colonial authors describe a rapid phase of imperial expansion by the last few Inca rulers, several chronicles offer a gradual perspective on Inca imperialism, generally describing a more steady imperial territorial expansion from the time of the founding Inca ancestor, Manco Capac.?® Until the 1940s, scholars such as Philip Ainsworth Means suggested that the story of a rapid Inca imperial expansion was a Spanish invention, devised to discredit the Incas and portray them as parvenu tyrants outside of Cusco. A series of influential publications in the 1940s by John H. Rowe reoriented the debate over Inca imperial expansion and firmly established the rapid expansionSTATE FORMATION, IMPERIAL EXPANSION 27 model as the one best supported by the available ethnohistoric and archaeological evidence from the Inca heartland and provinces. In 1945, Rowe observed that archaeological evidence, coupled with different kinds of reliable ethnohistoric sources, pointed to a rapid and late Inca imperial expansion. His systematic study of Inca political origins using archaeology and a critical reading of documents shifted the question of Inca political origins from a historical to an anthropological one. The German archaeologist Max Uhle had advocated a more active interpretive role for archaeology several decades earlier, but Rowe was the first to develop artifact chronologies in coordination with documentary research on the Incas.3° The rapid expansion paradigm for Inca imperialism still stands largely intact today, and most archaeologists agree with its general processes, if not the calendar dates presented in the Inca histories.?! In the 1940s, Rowe did not attempt to push the available data to resolve the issue of bow such rapid expansion was made possible, although he did suggest that some chronicles contained information on developmental processes after about AD 1200 and leading up to Inca imperial expansion.*2 Without a strong independent database from regional archaeology, the chronicle references to early developments in the Inca polity were considered too unreliable to be studied critically.33 Long-Term Change and Inca Origins Shifting the interpretive focus to processes of long-term change is the first step in moving beyond “great man” paradigms of Inca origins to develop more anthropological perspectives on the Inca empire. Scholars following the paradigm of rapid Inca imperial expansion have attempted to explain the conditions that enabled (or necessitated) such rapid, punc- tuated expansion. Factors hypothesized to have influenced the Inca developmental trajectory include institutionalized exchange relationships, internal conflict and interregional hostility, the centralization of economic management and redistribution, and ideological innovations.3+ Even in the absence of new data, significant scholarship has shifted the interpretive focus from singular events and the deeds of heroes and kings to the long-term development of social, economic, military, and ideological power. Like these scholars, many archaeologists tend to explain rapid Inca imperialism as the result of long-term regional processes of state formation and political consolidation.°>28 How the Incas Built Their Heartland Archaeological Inquiry in the Cusco Region In the nineteenth century, Cusco was a destination for many European travelers exploring South America, and authors such as Charles Wiener and Ephraim Squier provide descriptions of some of the region’s surviv- ing archaeological monuments.*¢ Aside from the acquisition of collections (probably from looted tombs), Max Uhle’s excavations at Q’atan near modern Urubamba, and Hiram Bingham’s work at Machu Picchu, little systematic excavation work was undertaken prior to the 1930s.>7 Archaeological research in the Cusco region began to flourish in the early 19308, due to an increase in government funding to celebrate the four hundredth anniversary of the Spanish founding of the city of Cusco.** At this time, Peruvian archaeologists began to conduct large-scale excavations at such monumental Inca sites as Saqsaywaman.?? The interpretive power of Cusco archaeology improved with the establishment of a preliminary artifact sequence by Rowe in 1944. Since the 1940s, the archaeological understanding of the pre-Columbian past has expanded as researchers have elaborated the artifact chronology and explored important time periods in the region. Large-scale excavations have been conducted throughout the region and have addressed the full range of human occupation in the Cusco region.*° The protection and study of monumental sites by the National Institute of Culture (INC) has led to the identification of hundreds of archaeological sites within the Inca heartland. Systematic Surveys in the Inca Heartland While excavation work has increased the interpretive power of relative artifact chronologies, regional systematic survey projects have begun to map changing settlement patterns in the Cusco region (fig. 2.1). Prior to the 1980s, numerous researchers had reconnoitered the region, making observations on the sites they encountered, but not systematically studying a defined region.*" The first systematic regional survey project was conducted in the Paruro area by Brian Bauer, who directed a pedestrian survey of over 600 square kilometers in a region just south of the Cusco Basin, registering over four hundred sites in a region where only a hand- ful had previously been known.*? Based on settlement pattern continu- ies in Paruro after AD 1000, Bauer concluded that an Inca state had formed in Cusco well before AD 1400, extending its control over theSTATE FORMATION, IMPERIAL EXPANSION 29 “Ancasm: as a ND Sea ar iz SS Nee 2 ee ac © Excavated Site |, Paruro Archaeological Project, directed by Brian Bauer Il. Cusco Valley Archaeological Project, directed by Bauer Ill, Sacred Valley Archaeological Project, directed by author Fig, 2.1. Important sites and survey projects in the Cusco region. Although excavations have been conducted in the region for nearly a century, it was only in the late 1980s that systematic regional settlement data made it possible to describe pre- Inca settlement patterns. Paruro region without substantial modification to local political and economic organization.4> Testing the hypothesis of early Inca state formation led Bauer to direct a second major regional survey in the Cusco Basin, a region in which approximately fourteen hundred sites were identified (fig. 2.2).*4 Settle- ment patterns in the Cusco Basin and to the east and west provided a second major region for comparing settlement change over time, and the(Aaorsipy emen jo umasnpy west -rauny ‘Sa0taiag Aresqry Jo “idaqq Asounog ‘9SZb€€ “ou aanefan) ‘of ‘UISeg ODS aL *t*STATE FORMATION, IMPERIAL EXPANSION 31 presence of an early Inca state has been substantiated with the new data.45 The same methodology and artifact sequence were used in both surveys, providing a foundation for expanding the regional database with subsequent survey projects.*6 The Sacred Valley Archaeological Project This book considers Bauer’s data from the Paruro and Cusco Basin surveys, but it also presents new data on over four hundred archaeological sites identified by the Sacred Valley Archaeological Project (SVAP), a survey and excavation project conducted in 2000 to expand the results of earlier survey research. The data from the Paruro and Cusco Basin projects had indicated that the Cusco polity developed early political control over the Cusco Basin and modestly populated areas to the south; assessing the extent and timing of this control and its relationship to the Spanish chronicles required that further research be initiated. The area selected for a third intensive settlement survey was a 30- kilometer stretch of the Vilcanota-Urubamba Valley, from just above the modern town of San Salvador to just below the town of Calca (fig. 2.3).47 In addition to the main river valley, the study region also included the side valleys draining into the Vilcanota River, a total area of approximately 400 square kilometers. This area was chosen for several reasons. From a practical perspective, the Sacred Valley study region abuts Bauer’s Cusco Basin survey, effectively connecting the three regional survey projects into a single 80-kilometer transect of the Inca heartland, a com- bined study region of about 1300 square kilometers in which over two. thousand archaeological sites have been located. The continuities in artifact types made it possible to employ a similar field methodology with an experienced crew of Peruvian archaeology students who had studied the Cusco ceramic chronology. Geographical and ecological considerations also influenced the definition of the study region. Because the aim of the project was to assess changing levels of influence and political control over the region, it was important to include areas that were close to the Cusco Basin, as well as lands that would have been beyond the administrative capabilities of an unspecialized government.*® The Chit’apampa Basin is located about a half day’s walk from Cusco, while walking to the Chongo and Qochoq valleys to the north of the Vileanota River would involve a much greater32, How the Incas Built Their Heartland , Patabamba@ oncocia -~ una Fig. 2.3. Modern towns in the Sacred Valley study region time and energy investment. Some sites in the Chongo Valley were over 25 kilometers from Cusco and would require over 2500 meters of ver tical climbing in each direction. The site of Ancasmarca (in the Qochoq Valley) was even more distant, with a similar elevation change. Based on environmental constraints, it was hypothesized that complex polities in the southern part of the survey region could be governed without the need for a specialized administrative hierarchy, but that the successfulSTATE FORMATION, IMPERIAL EXPANSION 33 incorporation of groups living to the north of the Vilcanota River would require state-level political organization. Ecologically, the study region was designated to focus on all areas of agricultural production, with small areas of high-elevation grassland (puna) included in the survey. Based on the traditional ecozone identifications described in the following chapter, the study region included areas of maize production in the lowest areas of the survey—from about 2900 meters around Calca—up to approximately 3600 meters in irrigated areas. Elevations from 3300 to 3800 meters are used today to dry- farm a wide range of crops. Above this is a zone of tuber production, and areas of grassy puna are found above 4000 meters. As defined, the study region included an elevation range of 2900-4500 meters, where a variety of human activities would have occurred in prehistory. Data on land use patterns, crop yields, and cultivation practices were available from ecological studies of the Vilcanota Valley and some side valleys in the survey region.4? One final consideration, and certainly not an insignificant one, was that the study region is mentioned frequently in the Spanish chronicles and colonial archival documents. Archaeologists had already identified and studied several Inca and pre-Inca sites in the region (fig. 2.4), including Ancasmarca, Juchuy Coscco, Warq’ana, Muyu Muyu, and Pukara Pantillijlla.5° Because earlier survey projects had identified where Inca state control was articulated early on—and without major disruption to existing settlement patterns—it was important to consider a region whose ethnic inhabitants had more complex, and often more contentious, interactions with the expanding Cusco state. Several Inca histories state that some local populations in the Sacred Valley were displaced during the Inca imperial period and their resources were developed as royal estates, four of which (Paullu, Caquia Xaquixaguana, Pisaq, Calca) were known to lie within the study area (fig. 2.5). In addition to regional survey work, the Sacred Valley Archaeologi- cal Project included excavations at the site of Pukara Pantillijlla, a large ridgetop village that was probably called Cuyomarca in Inca times.5" Five complete buildings were excavated at the site, as well as five test units, a total area of over 200 square meters that yielded a large sample of pottery and other artifacts. These excavations provide evidence of domestic occupation at the site prior to its incorporation into the Inca polity, as well as the period during which the site functioned as a secondary administrative site for the Inca state.34 How the Incas Built Their Heartland Je Ng Ancasmarka be WW) 4 km wv Archaeological site Site named in chronicles DE stachakancha Place named in age Ie ys eel Ee IRN) cayiomarca Conan agape re ws $3) KSiiane> Y, f ‘Juchuy Coscco/Warg'ana* wy. eae hs fr ) Pukara Pantillijlla Pisa 2.4. Sacred Valley sites known from the Spanish chronicles and previous archaeological research Summary As we have seen in this chapter, many factors make rapid imperial territorial expansion possible, bur one common feature of successful imperialism is the formation of a state that has begun to develop a centrally administered core region or heartland. Although long-distance conquests are known to precede the full administrative consolidation of this heartland, the state will have begun to unify previously independent groups-Caoistpy [eameN, Jo wunasnyy aneiso wou au, °S “ly ueapiony ‘saatasag Auvsqyy Jo “Idaqq Asausmnog ‘zoLPEE ‘ou anne BON) “br36 How the Incas Built Their Heartland under a centralized government that has a distinct ideology and shared ethnicity or citizenship. Within this region, the state will govern a large population and manage an intensified political economy, both of which provide the basis for a military force capable of, and ideologically in- clined to, a program of extended conquest. Many of the strategies for expansion and territorial administration developed during the state formation and regional consolidation process become the means for rapid imperial expansion. The proposition that state formation provided the means for rapid Inca imperial expansion not only accords with the regnant paradigm for rapid imperial expansion first articulated by Rowe but also offers an interpretive point of departure for the archaeological and historical study of Inca imperial origins. Regional settlement data and site-level excavations can be used to determine whether multiple indicators of Inca state formation are present before AD 1400, while a process-based reading of the Inca histories can identify some of the kinds of human interactions that would have occurred during the generations leading up to the first imperial campaigns. Although this book’s approach may depart from the methods and conclusions of some previous studies, it is consistent with the process- based approaches of other archaeologists and ethnohistorians. The collection of new archaeological data from the Sacred Valley study region permits a detailed analysis of the problem with reference to a huge vol- ume of previous archaeological and documentary studies. The new data confirm the propositions of many researchers as well as the expectations of anthropological archaeologists.CHAPTER THREE Ecology and Risk Reduction in the Inca Heartland In seeking to identify whether Inca state formation provided the basis for imperial expansion, we should also consider why a state would form in the Cusco Basin when it did, To do so, the process-based study of the ethnohistoric and archaeological data must be grounded in an understanding of the effects of environmental variability and climatic fluctu- ation on processes of culture change. Environmental conditions in the Cusco region today are highly variable, and human groups have pursued different strategies to manage environmental risks under specific ecological parameters. Certain conditions might seem amenable to the development or persistence of centralized states, but others would encourage strategies requiring less complex social and economic organization. In this chapter, the major ecological zones of the Cusco region are described, permitting us to consider two possible strategies that early sedentary farming groups could have used to reduce risk. Having looked at variations in the Cusco landscape, it is possible to address the question of climate change and its implications for human settlement patterns in the centuries leading up to Inca imperial expansion. Environmental Zones of the Cusco Region Many valuable studies of the ecology of the Cusco region have been published since the 1970s.! In addition, extensive ethnographic fieldwork has been conducted throughout the region, identifying local patterns of production and management of natural resources. In order to understand how the political economy of the Inca imperial heartland related to preimperial ethnic and ritual divisions, it is important to consider native ecological divisions within this large region. This book follows the typology employed by Jeffrey Parsons and colleagues, which is based 3738 How the Incas Built Their Heartland TABLE 3.1. Ecological Zones in the Cusco Region Ecozone Elevations and Human Use SVAP Area (%) Cordillera 4700-5700 meters; no economic use ° Upper puna 4200-4700 meters; pastoralism “5 Lower puna 3850-4200 meters; camelid pastoralism, tuber horticulture ~25 Upper kichwa (suni) 3500-3850 meters; mixed farming of tubers, ‘grains, tarwi 30 Lower kichwa 2700-3500 meters; maize agriculture ~40 Ceja de montana Below 2700 meters; cultivation of maize, ° tropical fruits, coca leaf on a number of classic geographical studies of Andean ecology.3 These researchers identify six principal ecological zones for the Andean highlands: cordillera, puna (separated into upper and lower subzones), kichwa (also separated into an upper subzone [sui] and a lower one), ceja de montana, montafa, and selva (or yunka). The most important ecozones of the Cusco region are the first four, and their approximate elevations and general production characteristics are listed in Table 3.1. Cordillera By definition, the cordillera lies above elevations where widespread human subsistence activities are common. In the Cusco region, many snow- capped mountains lying in this zone were venerated by the Incas and their neighbors, including Pitusiray, Sawasiray, Salcantay, and Ausan- gate.* Today, pilgrimages to the glaciers of Ausangate during the Qoyl- lur Rit’i festival mark important ritual congregations of groups from throughout the region, and the mountain of Pachatusan is part of an important pilgrimage from the Cusco Basin to the sanctuary of Wanka in the Sacred Valley.> Lower peaks without permanent snow cover (e.g., Pachatusan, Huanacaure) were also venerated regionally as sacred places (wak’akuna).© Areas of glaciated cordillera are found to the west of Cusco, as well as to the north of the Sacred Valley, although there are many other areas of high elevation, as seen in figure 3.1. The cordillera and upper puna surround relatively small valley areas where agriculture and herding can sustain large, dense populations. In Inca times the highest peaks wereECOLOGY AND RISK REDUCTION IN THE INCA HEARTLAND 39) Cordillera, over 4800 mas! Puna 4000-4400 mas! _ Pitusiray é a Qoricocha. a Salcantay e oO ..CUSCO et> 7 Pachatusai Huanacaure Below 4000 mas! " © oe ee eee 0 “ok Fig, 3.1. Areas of high-elevation grassland (puna) and cordillera in the Inca heartland used to delimit territorial boundaries (saywa) and were revered as the most important shrines in local and regional ritual life. Puna The puna zone in the Cusco region is not nearly as extensive as in the Titicaca Basin or Junin regions, but it did represent an important economic area in pre-Hispanic times. In the Cusco region, the upper puna (particularly elevations above 4200 meters) tends to consist of small spring-fed valleys and rolling high plains that would have provided40 How the Incas Built Their Heartland excellent pasture for camelid herds, as well as habitat for wild vicufias and deer species.” These areas are not productive agricultural zones, and many modern Andean herders living at such elevations make exchanges with agricultural communities at lower elevations. The viability of small family herds is maintained through social relationships that reduce risks of predation, theft, and herd disease.* The lower puna is used for grazing, as well as for tuber horticulture, with a wide variety of potato and other root crops (including oca, ullucu, mashua, and maca) typically cultivated in small, dispersed plots.? Above 4000 meters, only the hardiest, most frost-resistant crops thrive, and soils need long fallow periods between cultivations. Several researchers have noted the incredible diversity of tuber production in the Cusco region, often counting several dozen different varieties in a single community.!° Andean tuber production is aimed at averaging out ecological risks over time and space. The practice of freeze-drying root crops (especially bitter potatoes) allows families to manage annual crop variations, while maintaining biodiversity and dispersed plots reduces the risk of total crop failure in a given year.'! Ecological studies in the Department of Puno by Carol Goland have demonstrated that the effi- cacy of such risk reduction strategies declines as distance from the household increases. '? Puna grasslands are prevalent to the south of the Cusco Valley, as well as on both sides of the Sacred Valley. The Qoricocha area just north of Cusco is still an important herding area, and the chronicler Bernabé Cobo mentions that the Incas kept state herds there (fig. 3.2).'> Lower puna lands are common throughout the Cusco region and represent important agricultural resources for the production of tubers." Suni As with the lower puna elevations, lands in the upper kichwa (also known as suni) tend to be dry-farmed and are exploited in the same system of small, diversity-rich plots (fig. 3.3). Suni lands can often be used for more than one growing season but require an extended fallow period to recover. Within this zone, horticultural products include quinoa, kaftiwa, amaranth, tarwi, fiuhas (popping beans), and many of the root crops mentioned in the preceding. Studies in the Cusco region indicate that European cultigens like wheat, barley, broad beans, and onions have replaced many of the Andean crops, which are hardy and nutritious but-sseoedye pue scurvy jo spaay day axvy 01 uNoUy aze sora vaUT azyM are UE “pUNd eYDOI1IOD AY “TE “BLFig, 3.3 Modern dry-farming of upper kichwa plots. Beans, tubers, and other cultigens are intercropped in the same field.ECOLOGY AND RISK REDUCTION IN THE INCA HEARTLAND 43 Fig, 3.4. Suni and kichwa lands in the Inca heartland more difficult to process.'5 Suni lands dominate the sides of most valleys in the Cusco region, including much of the Anta-Maras-Chinchero area. Annual rainfall in the Cusco Basin and Anta-Maras-Chinchero area averages around 740 millimeters but is extremely variable from one year to another.'6 Lower Kichwa In distinction to the suni zone, Cusco’s lower kichwa elevations are dominated by intensive maize agriculture, particularly between 2800 and 3200 meters (fig. 3.4). Improved maize lands are the most productive agricultural resource in the region but have very limited distribution. The floor of the Cusco Basin is used for maize production, as are parts4 How the Incas Built Their Heartland of the Oropesa and Lucre basins. The most productive maize lands are found in the Vilcanota-Urubamba Valley, from below Quiquijana to around Ollantaytambo.'7 Below 3200 meters, the risk of frost is much lower, but these areas tend to receive much less rain than the upper kichwa lands around the Cusco Basin. Four towns in the prime maize- producing zone (Urubamba, Yucay, Calca, and Pisaq) receive an average of about 520 millimeters of annual precipitation, which is so variable that irrigation is necessary to ensure crop success. Intensive maize can be grown most successfully near the valley bot- toms throughout much of the Cusco region, and many narrow valleys like the Apurimac do not offer large areas for cultivation. Others have had to be improved to deal with seasonal flooding, meandering river beds, and steep valley sides (fig. 3.5). The early Inca state is thought to have executed much of this intensification. Ceja de Montafia In Cusco, ceja de montafia lands are restricted to the areas of Vil- cabamba, Lares, and Paucartambo. Traveling to these areas from Cusco is difficult, as the Vilcanota Valley and some major areas of cordillera must be traversed. This zone is warmer and more humid than the kichwa and can be used for the cultivation of crops prized by highland groups, such as coca leaf and chile peppers (chu). Some tropical fruits can be cultivated in this area as well, and some lowland varieties of maize flourish here.!8 Montana and Selva The montafia and selva ecozones did not form an important part of the Inca heartland. Indeed, the Incas viewed the tropical lowlands as a cul- turally foreign place that was dangerous and unhealthy for long-term occupation.'? In terms of agricultural products, these zones produce very different foods, including yuca, peanuts, and sweet potato. Natural and wild products like honey, feathers, and gold came from these areas, but the chronicles suggest that the Incas did not administer territory in the montafia and selva to the north and east of Cusco until the imperial period. ‘As with other parts of the Andean highlands, there is tremendous variability in climate, precipitation, and geology within the general elevation(daorsipy yeameNy Jo wnasnpy wesHIUEY ‘saoqasag Areaqyy Jo “1daq_ Asausnog *€9ZbEE “ou aanesoN) “besig we sosesI01 pur UoHEZITeUED 19ATy SE “Bl 19g Aawaqry Jo ad0qq 0 ‘£9. NeBON) ‘best pur vonezyt ny SE By46 How the Incas Built Their Heartland zones described.2° As a region, Cusco has a diverse distribution of agro- pastoral resources, but its modern economic base is characterized by kichwa groups who coordinate the production of diverse resources. Contemporary communities near Cusco use a variety of means to ex- ploit a wide range of ecological zones, including social relationships, archipelagic settlement patterns, and market exchange.2! Two Strategies of Agricultural Risk Management Understanding the Inca state as a kichwa-based polity raises the issue of political economy and strategies for risk reduction. The simplest distinction to make in discussing this is between dry-farming and the pur- suit of more intensive agricultural methods, namely irrigation, artificial fertilization, and improvement of farmland.22 Dry-farming of a wide variety of crops is possible on most lands in the Cusco region, while intensive irrigation-based agriculture is much more restricted. The risk- reduction benefits of dry-farming can be realized with relatively low population densities, the maintenance of social relationships, and a relatively simple political economy. Intensive agriculture reduces risk through surplus production, storage, and improved techniques, permitting higher population densities but generally requiring more centralized or complex management. It is important to point out that these strategies are not mutually exclusive. State political economies might focus on intensive agriculture while subsistence farmers practice dry-farming, even though they might prefer more intensive methods. The Sacred Valley study region is characterized by two main economic production areas, Improved valley-bottom lands represent a small percentage of agricultural lands, but they yield huge agricultural surpluses, almost exclusively maize. Many of these lands were developed by the Incas and have remained as private properties in the hands of Cusco’s elite until the present. In the side valleys that drain into the Vil- canota-Urubamba River, some irrigation is possible at lower elevations, but most agricultural production consists of dry-farming multiple plots in upper kichwa and suni lands, where a diverse array of products is grown. Dry-farming is pursued on the vast majority of agricultural lands in the Cusco region, but such resources must be rotated and fallowed to ensure soil quality. Most dry-farmed lands are now held corporately by peasant communities who manage land use at the community level.ECOLOGY AND RISK REDUCTION IN THE INCA HEARTLAND 47 Dry-Farming as Subsistence Agriculture Ideally, subsistence agriculture focuses on satisfizing rather than the production of surplus—that is, meeting household food needs with a minimum of labor input and risk.23 Household production cannot simply track anticipated needs, however; it must also cope with possible fluctuations in productivity caused by climatic variation.24 For small groups of sedentary farmers, this has traditionally involved a focus on crop diversity and maintenance of multiple small plots as a means of buffering the risks of crop failure and starvation.25 Temperature, precipitation, and other climatic factors fluctuate widely throughout a given region, as well as from year to year, creating a patchy resource base that is un- predictable from one year to the next.2 To minimize risk, Andean peasants intercrop multiple varieties and species in the same plot and maintain multiple plots in different locations.27 Crop rotation and fallowing are necessary to maintain soil quality, and modern peasant communities manage such practices locally. Farming multiple plots in a mountain environment means that in a given year at least some plots should be productive, but cultivating diversity also incurs travel costs, thereby reducing net caloric yield of a given crop.?8 Generally speaking, the proximity of land resources sets effective limits for diversity, and Andean peasants must balance risk management with potential declines in net caloric yield, rejecting the expansion of diet breadth when it would impose negative (or risky) net returns.2? Crop diversity, while an important risk-management strategy, is per- ceived by peasant farmers as the means to satisfying a “complete” diet breadth. Working in the Paucartambo Valley, Karl Zimmerer has shown how food is used to shape cultural identity and status, noting a recent trend among Andean peasants to value native cultigens over European ones.?° It is clear that culinary values and considerations of secondary production costs (e.g., processing, requirements of fertilization, weed- ing, etc.) influence decisions of what to cultivate.3! A complex set of criteria determines the cultural value of a given species or variety, and it has been observed that sometimes the cultivation of rare or low-yielding varieties is undertaken because of their high status as gifts. Based on settlement locations, many pre-Hispanic agricultural groups in the Cusco region probably relied primarily on agroecological diversity to reduce risk, cultivating several kinds of crops to maintain a complete diet (fig. 3.6). For groups engaged in both farming and herding, the48 How the Incas Built Their Heartland principal settlements would tend to be located at the transitional elevations between production zones in order to maximize diversity while minimiz- ing travel costs. Even after the introduction of intensive maize or tuber agriculture as a production strategy, diversity-rich farming has continued in one form or another until the present.3 Principal carbohydrate sources for Andean farming groups include maize and root crops such as potatoes (Solanum spp.), oca (Oxalis tuberosa), ullucu (Ullucus tuberosus), mashua (or aiiu, Tropaeolum tuberosum), and possibly maca (Lepidium meyenii), many of which are still cultivated in the Cusco area.>4 While maize hasa limited distribution in the Sacred Valley—typically in irrigated fields below 3500 meters—different varieties of tuber crops can be dry- farmed at elevations up to 4300 meters and are still grown today in an astonishing diversity. As Robert Rhoades and Virginia Nazarea observe: Farmers opt for an adaptive strategy of using biodiversity in such a way that it spreads production risk and labor scheduling across the landscape. In the Cusco Valley of Peru, for example, we found farmers who plant up to 50 varieties as well as several different species of potatoes at different time intervals in 20 to 30 scattered fields characterized by different altitude, soil types, and orientations to the sun. . . . This dispersion pattern reduces the risk that one disease outbreak or an unpredicted frost will devastate an entire crop. Simultaneously, by using different varieties a continuous flow of production through time and space can be realized so that different markets, household needs, or labor supplies can be accommodated. Interspecific and intraspecific variation is also used for agronomic control of weeds and pests, micro- climatic variation through shading, as well as a buffer against climatic and pest damage. Andean potato farmers’ strategies are based on a long-term, detailed knowledge of specific plant-environment interaction. Any variety is tested against several seasons of variable frosts and rainfall as well as performance in different soils.35 Studies in the Paucartambo Valley and our own field observations in the upper basins of the Sacred Valley confirm that diverse cropping of tubers constitutes a major food source for subsistence farmers.°¢ While European cultigens such as wheat, barley, and broad beans have largely replaced Andean equivalents in places like the Sacred Val- ley, quinoa (Chenopodium quinoa) and amaranth (Amaranthus cauda- tus) are still cultivated in many places.*” Kajtitva (Chenopodium pallidi- caule) might also have been an important crop in pre-Inca times, when populations shifted to high elevations and had to contend with colderTRAVAXO dezienGu- capac yobray mi Jecdimbea Fig. 3.6. Andean peasants sowing potatoes and oca. (Redrawn from Guaman Poma de Ayala 1980 [16r5].)50, How the Incas Built Their Heartland temperatures and shortened growing seasons. These plants provide very high percentages of high-quality proteins, as well as significant amounts of carbohydrates and some fat.3* The Andean grains are frost and drought resistant and are thus less risky to cultivate on more marginal lands, although they require more time to process for consumption. Today, a common crop rotation pattern is tubers-quinoa-tarwi.>? Tarwi (Lupinus mutabilis) has an extremely high protein content (averaging 46 percent), as well as oil and fat (about 20 percent), and as a nitrogen-fixing plant can be cultivated at the end of a rotation cycle, intercropped to protect other plants, or grown on extremely marginal lands to improve them for other crops.*° Tarwi is still grown in the Cusco region and provides an important supplement of protein and fat for the subsistence farmer. In describing the cultivation of diverse products as a risk-reduction strategy, we should recall that Cusco’s agricultural groups were not strict satisfizers. The earliest sedentary villages reveal evidence of long-distance exchanges of goods not available locally, such as decorated pottery and obsidian. Exotic goods with utilitarian, status, and ritual uses could have been acquired through exchanges or by obtaining them directly by traveling to where they were produced or occurred naturally. Direct exploitation and exchange both would require some sort of surplus production, and local pre-Inca elites would have been involved in acquiring exotic goods, or at least in providing leadership and organization for the economic and social interactions necessary to acquire them. While still managing production systems in areas best suited for diverse dry-farming, some pre-Inca elites also mobilized local labor to build canals and small terrace systems. The formation of the Inca state shifted the focus of agricultural production from upper kichwa lands to new state lands in the lower kichwa, where intensive maize production would have radically altered local economies and diets. Intensive Agriculture and the Inca Empire Paul Halstead and John O'Shea have observed that in addition to diversi- fication, human groups can also address resource fluctuations through mobility, storage, and exchange.*! Rather than cultivate biodiversity, the Inca empire chose to move populations to resources (mobility) and construct storage facilities, while simultaneously intensifying agricultural production (fig. 3.7). The intensive agriculture favored by the Inca stateBINS = oh ~ a 3 Ae eX 3 ea é Ss é REZ \Ws Ls y zg EW, ISS") = See 352, How the Incas Built Their Heartland used irrigated, well-drained, terraced lands and labor tribute to produce surpluses of a limited array of crops. As Zimmerer notes, this state agriculture was probably diversity-poor, focusing on a few varieties that would meet limited state functions.4” The most important crop was maize, the cultivation of which was a central concern of the Inca religious calendar.* Several chroniclers attest to the central role of maize and its almost exclusive cultivation on irrigated and terraced lands.*4 ‘The Incas intensified local maize production, building irrigated terrace systems and scheduling planting, irrigation, harvest, and fertilization. While local lands were sometimes annexed for state use, the canalization of rivers, draining of swampy land, and introduction of a reliable water supply created agricultural resources that not only were less risky than dry-farmed fields but also did not require a long fallowing between plantings.45 Fertilization with camelid dung, night soil, and guano from the coast raised the probability that these fields would yield large surpluses year after year.4® The state focus was on a few maize varieties, including one that would store well, such as muruchhu sara, a flinty maize said to have been preferred by the military.47 In addition to addressing long-term storage concerns, the Incas developed high-yield varieties like Cusco Cristal Amarillo, which originated in the Sacred Valley and was introduced in Inca provinces to replace local strains.** Finally, the state required a few varieties (such as Cusco Cristal Amarillo and Chullpi) that had high sugar content, suitable for making maize beer (agha).4? In places where maize was not the dominant staple crop (e.g., the Aymara-speaking Titicaca Basin), tuber production was reorganized by the Inca empire.°° Of the estimated six thousand varieties of potato available in South America, the Incas selected a few for high yield, as well as for making ch’ufiu, a freeze-dried product that could be stored for long periods of time and easily transported.>'! While Ludovico Berto- nio’s Aymara dictionary lists perhaps twenty varieties of potatoes and tubers, Diego Gonzalez Holguin’s Quechua dictionary provides limited vocabulary for tuber production, almost all of it ch’ufiu related.>? In the Cusco region, the Incas sowed early (away) varieties of both maize and potatoes, crops that could be harvested several months before normal harvests would be ready.°> Inca surplus food production focused primarily on maize and ch'wiiu, but these were not the only agricultural products grown on improved state lands. Archival documents and chronicle references indicate thatECOLOGY AND RISK REDUCTION IN THE INCA HEARTLAND 53 coca leaf, chile pepper (uchu), cotton, and other crops were cultivated in lower-elevation state fields. # In some cases, terrace systems appear to. have raised field temperatures sufficiently to raise exotic crops on state lands. These products represent the development of agricultural wealth and are discussed in greater detail in chapter 10. Climatic Change in the Andean Highlands Agricultural strategies of resource diversity or surplus production are intended to address interannual fluctuations in precipitation and temperature. Both strategies could survive periodic resource failure, but it is likely that sustained climatic shifts would require a revision of risk- management strategies. Although climate change should not be promoted as the prime mover in long-term culture change, it would have had varying effects on groups practicing different political economies.55 Several indicators of prehistoric climate change have been measured for the An- dean highlands, and some general trends can be noted for the past 1500 years or so. In the Cusco region, the Marcacocha lake core and Quelc- caya ice core data provide comparable evidence for long-term climatic change that can be compared to other paleoclimate studies.°¢ The Marcacocha sequence shows a change in lake bed deposits around AD 100, interpreted as a cooler period from AD 100 to AD 1050.57 An earlier stratum contained pollen possibly associated with the dry-farming of hardy crops such as quinoa, kafiiwa, and maca, but this pollen de- creases from AD 100 to AD 1050. The lake core evidence suggests the onset of variable climatic conditions after AD 900, with warmer and drier conditions than before.>® The cultivation of quinoa and maize, and possibly tubers like potato and oca, would have been undertaken, and the construction of irrigation systems using melt water developed from AD 1100 to AD 1500.5? Ice core data from the Quelccaya glacier (located about 100 kilometers east of Cusco) indicate similar trends in temperature and precipitation from ca. AD 540 to AD 1984 (table 3.2).5° The period from AD 540 to AD 760 was characterized by generally drier conditions, including marked dry periods from AD 540 to AD 560 and 650 to 730 and a severe drought from AD 570 to AD 610.5! Only the period from AD 610 to AD 650 was wetter than average. This period of warm, dry conditions was followed by a prolonged cool period from AD 760 to AD 1040,54 How the Incas Built Their Heartland TABLE 3.2. Wet and Dry Periods in Quelecaya Glacier Cores Wetter Periods Drier Periods AD 1870-1984 1720-1860" 1500-1720" 1250-1310" 760-1040 650-730 610-650 570-610" 540-560 Source: Thompson etal. 1985, 973. Extremes, in which precipitation deviates more than 20 percent from the mean, in which increased precipitation was common. Warm, dry conditions prevailed from AD 1040 to AD 1490, with what L. G. Thompson and colleagues call an “especially intense” dry period from AD 1250 to AD 1310.2 Using data from the Nevado Huaytapallana glacier, Geoffrey Seltzer and Christine Hastorf interpret glacial advances at this time as a sign that temperatures at the end of the thirteenth century were cooler than in preceding centuries.°? They estimate that climatic zones would have been depressed by as much as 150 meters, radically altering the productive landscape in highland valleys. The onset of the Little Ice Age around AD 1500 brought cooler and markedly wetter conditions that would last until the eighteenth century.°*+ Settlement Patterns, Climate, and Political Economy Paleoclimate data provide a means for approaching some of the issues of settlement location and political economy. Major climatic shifts would have altered the costs and benefits for certain settlement locations and production strategies. Warmer temperatures would have raised the elevational limits for cultivating specific crops, while conditions of increased aridity generally associated with such changes would have favored the cultivation of hardy, drought- and frost-resistant plants such as potatoes, oca, ullucu, quinoa, and kafiiwa.% In times of drought, intensive valley- bottom agriculture would continue to be a successful production strategy in locations with permanent streams, provided that intensification did not outstrip reduced water resources.6° Cooler conditions would tend to favor shifts to lower-elevation settlement as the limits of ecolog-ECOLOGY AND RISK REDUCTION IN THE INCA HEARTLAND 55 ical zones were lowered. The concomitant increase in available water might encourage the expansion of intensive agriculture into more marginal environments, and such periods might be characterized by an increase in maize cultivation. In purely environmental terms, diversity and intensive production both represent viable means of addressing interannual environmental fluctuations.°7 Major climatic shifts would affect these economies dif- ferently. A sudden shift from wet/cool to dry/warm conditions would have a negative impact on large systems of intensive agriculture, as has been argued for the Tiwanaku empire. Sudden climate change or cata- strophic environmental perturbations such as earthquakes or El Nifio Seasonal Oscillation (ENSO) events can have a devastating effect on centralized polities whose economies rely heavily on intensified agricultural resources. Conversely, a shift from dry/warm conditions to wet/cool ones would affect high-elevation settlement systems relying on diverse cropping, as lowered elevational limits for crops (and too much precipitation) would impose greater travel costs and more frequent crop failure, reducing net returns from cultivation. Major climatic changes stress social systems by reducing food resources, but they also reduce confi- dence in the political and religious systems in which affected groups participate. The settlement patterns in the Sacred Valley exhibit shifts in site location, only some of which can be accounted for using paleoclimate data. The first villages in the study region tend to be located in the suni zone, in areas where herding and dry-farming of diverse crops could be managed, probably organized at the level of the village or small polity. At around AD 400, the large, nucleated villages in the Sacred Valley study region were abandoned as population shifted to locations where simple irrigation systems could be exploited, possibly for an increased production of maize. These settlement patterns remained more or less stable for about 600 years until around AD 1ooo, when a second shift to nucleated high-elevation settlements occurred. This was followed by a gradual resettlement of population closer to the valley bottom beginning around AD 1300-1400. Many of these settlement shifts reflect changes in political or economic organization, and the following chapters consider how regional ecological changes influenced the formation of the Inca state.CHAPTER FOUR Wari Imperialism and Local Political Competition in Cusco Some of the first Spaniards to traverse the Andean highlands described the ruins of impressive sites that had been built before the rise of the Inca empire.' Although the Incas insisted that they had been the first civilization to develop in the Andes, archaeologists now know of other, earlier, states and empires. Nearly a century ago, the archaeologist Max Uhle observed the need to link the Inca empire to these earlier polities using archaeological data. Some scholars look to the Wari empire as the inspiration for Inca statecraft.? This state flourished in the central Andean highlands and some coastal regions from around AD 600 to AD 900 or so and is known to have occupied and influenced the Cusco region, although consensus has not been reached regarding the degree to which Wari colonization influenced Inca state formation. Some researchers suggest that the Incas inherited the Wari model of statecraft, preserved by the descendants of Wari colonists in the Cusco region in the centuries following the abandonment of Wari sites and leading up to Inca imperial expansion.? Although the Inca and Wari polities did share some structural similarities—for example, an urban capital, standardized state architecture, and the development of roads and intensive maize agriculture to facilitate state functions—they appear to have exercised different levels of intensity in provincial administration and the central direction of political economy. The development of regional settlement patterns now allows us to discuss the effects of Wari imperialism on political hierarchies and competition in the Cusco region. The available evidence suggests that a complex prestate polity already existed in the Cusco Basin when Wari colonies were established. This polity was not heavily impacted by the Wari presence, but centuries of contact with Wari colonists and administrators do appear to have transformed the nature of elite identity and group competition. 56WARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 57 © capital A Wari site I Tiwanaku site __ Tiwanaku >>~Heartland >. N. as Azapa Area of Tiwanaku Influence . 4ot. Estimated territories of the Wari and Tiwanaku polities. These states did not exercise continuous control in the regions under their influence. Circles mark the locations of the two capitals, and triangles and squares identify important Wari and Tiwanaku sites, respectively. Wari and Tiwanaku Integration of the Andean Region ‘The first Andean states formed on Peru’s coast no later than AD 300, and succeeding cycles of state formation and regional balkanization extended the scope of state administration into highland regions.4 Between about AD 500 and AD 700, the Wari and Tiwanaku polities united their respective highland core areas politically and began to establish colonies and provinces in many highland and coastal areas (fig. 4.1). The Cusco region has a marked presence of Wari colonial settlement that has been58 How the Incas Built Their Heartland well studied, and the decline of Wari colonies there set the stage for Inca state formation and imperialism. Analysis of recently collected archaeological evidence from the Cusco region suggests that although the Wari colonial presence influenced Inca state formation, it was in less direct ways than some researchers have assumed. Recent reviews of early Andean empires have pointed out that administrative control implemented outside of the Wari and Tiwanaku heartlands was territorially discontinuous and many “provincial” regions lack evidence of infrastructural investments consistent with the imposition of direct rule.‘ It is clear, however, that these polities achieved a degree of macroregional integration—particularly in the formal linking of coastal and highland regions—that was beyond that of early coastal polities like the Moche state. Both Wari and Tiwanaku polities had urban capitals surrounded by densely populated hinterlands.6 Wari roads and rectangular architectural compounds have been identified widely throughout the Andean highlands, but the best evidence of resettlement and economic reorganization consistent with direct state administration is found within 100 kilometers or so of the capital.” The Tiwanaku capital administered a heartland of comparable size. As observed in chapter 2, territories of 20,000 square kilometers or more are comparable to those observed archaeologically for many other early states. Outside of its heartland, the Wari polity established its presence widely throughout the Andean highlands by establishing colonies in strategic locations.? The Wari appear to have had direct access to some coastal valleys that could provide a wide range of resources not available locally, including guano, seaweed, coca leaf, Spondylus shell, and fish."° In many regions colonized by Wari settlers, foreign influence was limited to a very small area around the colonies, with almost no evidence of direct interactions between local populations and the state heartland.'! The Tiwanaku polity intensified the agricultural resources of its heartland, constructing raised fields near the shores of Lake Titicaca at over 3800 meters. By draining large expanses of new agricultural land, Ti- wanaku farmers were able to cultivate tubers intensively with lower risks of frost, providing food for hundreds of thousands of people living, in and around the capital.!2 While surpluses of staple foods were produced abundantly, ceremonial crops like chile pepper and coca leaf could not be cultivated in the Titicaca Basin. Tiwanaku established formal colonies at lower elevations to the east and west of its heartland: in the piedmont lands of some Pacific Coast valleys (Moquegua, Sama, andWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 59 Azapa), as well as the Larecaja and Cochabamba regions in the Bolivian lowland valleys.!3 These colonies afforded direct access to maize, chile peppers, and coca leaf, as well as exotic goods acquired through contacts with groups living on the Pacific Coast and in the Amazonian Basin.'4 Although the Wari and Tiwanaku polities might not have exercised direct control over all resources used in their political economies, they should still be considered to have been empires on the basis of having, established colonies and administered ethnic groups living beyond their respective heartlands. The transition from colonization to provincial administration appears to have been a long-term goal of these polities, and formal provincial administration was either present or under development in some areas at the time that these polities declined.'S Local Political Organization in the Cusco Region, AD 400-600 Recent settlement surveys conducted in the Cusco region indicate that a complex polity already existed in the nearby Cusco Basin at the time that Wari colonists first began to settle in the Lucre Basin. Researchers have identified a local pottery style called Qotakalli that was produced prior to Wari imperial expansion.'6 Qotakalli pottery appears to have been produced in the Cusco Basin by about AD 400 and was distributed within that basin and in several neighboring regions.'7 This local style continued to be produced during the period of Wari colonization (after about AD 600), but a new local style called Araway—a local imitation of Wari imperial pottery made with a similar paste to that used for the Qotakalli style—came to be the more common decorated ware."® The distribution of Qotakalli pottery suggests the presence of a complex prestate polity in the Cusco region that might have controlled an area of up to 1000 square kilometers. The Cusco Basin Polity In the Cusco Basin, Brian Bauer has identified more than a dozen vil- ages, many of which form distinct clusters on low ridges just above the valley floor (fig. 4.2).! Because of the extent of modern settlement in the valley and the more dispersed nature of Qotakalli Period settlement, itis difficult to identify a paramount center, although the argument could be made that the center of the Cusco polity comprised a cluster of villages60 How the Incas Built Their Heartland <<" Chit'apampa’_ 4000S, Basin © SR TES * redrawn from Bauer 2004 ==) @ .25-1.00 ha. ’ 7 rim ee Fig. 4.2. Qotakalli sites in the Cusco Basin (after AD 400) located near the western end of the basin. Prior to the Wari occupation of the Lucre-Huaro area, the Cusco Basin was characterized by a two- or three-tier settlement hierarchy, with a few clusters of two or three small villages (1-5 hectares each) possibly indicating the most important settlement areas. In addition to these are single small villages, some of them surrounded by hamlets, as well as groups of hamlets. Most sites are located in places that are better suited to farming than community defense, and the distribution of Qotakalli pottery to the level of hamlets suggests a high degree of interaction between settlements within the Cusco Basin.?° Limits of Cusco Control in the Sacred Valley Settlement patterns from the Sacred Valley Archaeological Project reveal a major settlement shift in the Chit’apampa Basin—located just over a low pass from the Cusco Basin—at around AD 4oo. For several hundred years prior to that time, settlement in this basin had been characterizedWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 61 CHIT'APAMPA BASIN Formative Sites ca. 1000 BC- AD 400 @ Under 1 ha. I Small Village (1-5 ha) e Large Village (>5 ha.) CUSCO BASIN Fig. 4.3. Formative sites in the Chit’'apampa Basin (ca. 1000 BC-AD 400) by a three-tier settlement hierarchy, with the largest villages located about 15 kilometers from Cusco on lands where dry-farming could be pursued (fig. 4.3). This settlement pattern is consistent with a small chiefdom, an independent polity whose elite leaders probably had regular contacts with populations living in the Cusco Valley. The largest early villages of the Chit’apampa Basin cluster around Ragqchi, a site built on a promi- nent ridge overlooking an important route from the Cusco Basin to the Sacred Valley (fig. 4.4). For the Formative Period, Bauer has identified another three-tier settlement system in the Cusco Basin, as well as thirty- one small sites in the Paruro region that may or may not have been organized hierarchically.2" At around AD 400, all large villages in the Chit’apampa Basin wereoot (ly punose pouopuege afeys aaneuuog doispu afzel v ‘typbey “bb “BryWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 63 CHIT'APAMPA BASIN Qotakalli Period ca. AD 400-600 @ Under 1 ha. I Small Village (1-5 ha) Fig, 4.5- Qotakalli sites in the Chit'apampa Basin abandoned, and a new settlement pattern was established at the lowest elevations of that basin (about 3500 meters) (fig. 4.5). In the Sacred Val- ley study region more than 70 percent of sites (thirty-eight of fifty-four) occupied from around 1000 BC-AD 400 have no evidence of continued settlement after AD 400. The new settlement pattern at that time is less hierarchical, and almost all decorated pottery found at the new Qotakalli Period villages in the Chit’apampa Basin was Qotakalli pottery from the Cusco Basin. The distribution of the Qotakalli pottery style in the Sacred Valley study region was quite limited geographically (fig. 4.6). Almost all of the sites with high percentages of Qotakalli pottery on the surface relative64 How the Incas Built Their Heartland @ Under ha. Ei Small village (1-5 ha) Fig. 4.6. Qotakalli sites in the Sacred Valley study region to other contemporaneous pottery styles are found in the Chit’apampa Basin. A cluster of twenty-three sites in the upper Chit'apampa Basin represents over 6 hectares of occupational area, comprising half of all sites in the Sacred Valley study region with this style present (fig. 4.7). As with the Cusco Basin, most of these are small hamlets, but several small villages of about a hectare in size (VS-082, VS-043, VS-o5 5) were also encountered. Because all of these sites were found in areas with substantial modern agriculture and settlement, the original size of these sites was probably somewhat larger. Two other areas with substantial Qo- takalli surface pottery can also be noted. Around modern Patabamba,spareatyno aq joneaayp saddn ayp sau porrag period passadsip yo vary °Z+ “Bg66 How the Incas Built Their Heartland four sites (VS-378, VS-379, VS-380, VS-381) indicate the presence of a small village and some outlying hamlets, a total occupation area of around 3 hectares. The other important site, Willkarayan (VS-130), is located in the lower part of the Chit’apampa Basin. This site, built on a low prominence overlooking the principal route between the Sacred Valley and the Chit’apampa Basin, was about 1 hectare in area, surrounded by concentric terrace walls. This site, together with the sites in Patabamba, marks the limit of sites where the Qotakalli style constitutes a distinct majority of decorated shards. The distribution of Qotakalli pottery drops off sharply to the north of the Vilcanota River, as well as at sites in the valley proper. Other sites with Qotakalli pottery are found in the Sacred Valley, as well as in its larger side valleys. The sites in the main valley tend to be quite small and are usually located 200 to 300 meters above the valley floor, in areas with natural defense that are close to small streams. Larger sites are found in the area of San Salvador around the Sanctuario de Huanea (Wanka), as well as in the Chongo Basin and around Cala. Qotakalli does not predominate at these sites, and the presence of local imitations of Qotakalli and Araway, as well as Wari Period styles (such as Wari and an unidentified style that may originate in the Paucartambo region), suggests that some of these sites continued to be occupied into the Wari Period. Regardless of the later occupation, it is clear that the Cusco styles comprise the majority of decorated surface shards only in areas within about a half day’s walk of Cusco. The distribution of the Qotakalli style in the study region reveals some important characteristics about the size and complexity of the Cusco Valley polity at this time. First, it is clear that Qotakalli dominates surface assemblages in areas close to the Cusco Basin, while the presence of the style drops off sharply to the north of the Vilcanota River. This comes as no surprise for a nonstate polity, whose leadership would have difficulties with the long-term administration of areas beyond about a half day’s walk from Cusco. Clear Qotakalli settlement clusters were established around the modern communities of Huillcapata and Se- queracay in the Chit’apampa Basin and Patabamba in the upper elevations of the southern rim of the main valley. The site of Willkarayan marks the limit of Qotakalli stylistic dominance along a major route into the Sacred Valley but does not necessarily represent a site under the control of a Cusco Basin polity. In the Chit’apampa Basin, the shift away from a pattern of nucleatedWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 67 settlements in upper kichwa lands (characteristic of the period from ca. 1000 BC to AD 400) to a more dispersed settlement pattern close to water sources is consistent with changes in local autonomy and economy. While it is difficult to say exactly how this area came to be dominated by groups living in the Cusco Basin, it is likely that interaction with these groups promoted new settlement patterns. In terms of defense, participation in a larger regional polity would reduce the dangers of raiding—and some of the need to live in nucleated settlements—while the leaders of the Cusco Basin polity would be expected to encourage movement away from sites affording protection from the paramount leadership. The shift toward water sources and agricultural lands may have been a response to tributary demands from Cusco, allowing locals to reduce travel time to agricultural plots in a safer political environment. Conversely, this may also be due in part to shifting production strategies, as higher-elevation lands were abandoned in favor of lands where simple irrigation systems could be developed. The cultivation of irrigated lands around 3500 meters may be indicative of an intensification of maize agriculture, although additional excavations are needed to confirm this. As mentioned before, the rise of political complexity in Cusco would have led to an increased participation in interregional exchange networks, and this may in turn have led to an increase in maize cultivation. The regional settlement patterns of the three Cusco archaeological surveys indicate that a complex prestate polity had developed in the Cusco Valley by the latter half of the Formative Period, spreading its influence throughout the area within about a half-day’s walk from a principal cluster of important villages located in the western end of the Cusco Basin. A general abandonment of Formative Period villages can be observed throughout the region, as well as the establishment of smaller settlements closer to valley-bottom agricultural resources. At this time, Cusco elites were probably in contact with emerging complex polities in other parts of the southern and central Andean highlands, and this may have led to the subordination of neighboring valleys, as well as a shift toward maize agriculture. Cusco Influence in the Paruro Region To the south of Cusco, Bauer has noted a similar geographical limit for Qotakalli pottery.2? This style was identified at nineteen sites in Bauer’s68 How the Incas Built Their Heartland Paruro survey area, sixteen of them located to the north of the Apuri- mac River, Almost all sites within 15-20 kilometers of the principal Qo- takalli villages in the Cusco Basin had high percentages of Qotakalli pottery, and 79 percent of all sites (fifteen of nineteen) with the style present were found within 25 kilometers of the principal Cusco Basin settlements. Bauer also identified a contemporaneous local style (Ccoipa) in the Paruro region, present in very small amounts within about ro kilometers of the main Cusco Basin sites but identified at thirty-five sites in the Paruro survey region.23 Sites with Ccoipa pottery were clustered in the small valleys on both sides of the Apurimac River, with 74 percent (twenty-six of thirty-five) found 20-40 kilometers from the principal Qotakalli villages, beyond the administrative limits of a nonstate polity. The distribution of Qotakalli pottery to the north and south of the Cusco Basin may not be a good marker of political control, but it does suggest marked differences in the rate of interaction and exchange between groups as distances from the Cusco Basin increase. Settlement patterns from the Sacred Valley and Paruro study regions indicate that the polity in the Cusco Basin was capable of dominating an area within about 20 kilometers of its principal settlements (fig. 4.8). Groups living outside of that area interacted with the Cusco polity but probably were not under its control. These included autonomous villages, as well as other small complex polities. Given these territorial limits, it appears that Wari established its colonies at the margins of the complex Cusco Basin polity.24 While Wari colonization left an indelible mark on the Lucre Basin and Huaro area, it did not disrupt settlement in areas where Qotakalli pottery was most prevalent. Archaeological Manifestations of Wari Imperialism in the Cusco Region The Wari empire established colonies in the Cusco region during the seventh century, at the time of its first major wave of expansion throughout the central Andes.?5 Archaeological research has identified two major Wari installations and associated settlement clusters located about 25-45 kilometers to the southeast of modern Cusco, at Pikillacta in the Lucre Basin, and at Batan Urqu located near modern Huaro.?° Some authors assert that religious patronage facilitated Wari colonization in Cusco, stating that Wari immigrants settled on or near important localWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 69 OZ 0 10 20km Fig. 4.8, Estimated limits of the territorial control of the Cusco Basin polity, AD 400- tooo. Pottery distributions indicate that this polity was not organized as a centralized state. shrines, co-opting sacred power and perhaps introducing the Wari state religion to local groups.?” According to Julinho Zapata Rodriguez, Wari burials have been found at Batan Urqu, a site that had a pre-Wari religious complex, while Gordon McEwan suggests that the site of Mama Qolla in the Lucre Basin was an important shrine co-opted by Wari settlers.?® The data for dating the establishment of Wari colonies in Cusco and assessing the regional importance of pre-Wari shrines have not been fully developed, so we must treat such hypotheses as preliminary. The Wari colonies in the Cusco region were established in locations where it would have been easy to monitor caravan traffic throughout theJO. How the Incas Built Their Heartland region, particularly along the Vilcanota Valley and out of the Cusco Basin. It is possible that the Wari state was also interested in access to lowland areas northeast of Cusco, where coca leaf and gold could be found, and there is evidence that raw materials like obsidian and turquoise may have been worked in the Cusco region.2° Wari colonists were probably able to attract local support through gifts of craft goods, obsidian, fancy pottery, and metal, which may have been given to reward labor service. Exotic goods available through Wari colonists would have presented co- operative local elites with the religious and economic means for ag- grandizing themselves, and exchanges with local groups may have helped Wari colonists to organize the labor that was invested in the ongoing construction of colonial infrastructure, including the site of Pikillacta. Pikillacta and the Lucre Basin Pikillacta is a massive installation comprising a 47-hectare rectangular enclosure (745 x 630 meters) and two areas of possible corrals in a total area estimated at nearly 2 square kilometers (fig. 4.9).2° Construction of the principal compound began around AD 600 and continued in three discrete phases. Some parts of the site were occupied until AD 900 or later, although excavations have revealed that the site was still under construction at the time of its abandonment and was never fully occupied.*! The main compounds at Pikillacta included open spaces for staging public feasts and more private ceremonies, as well as residential areas.32 The presence of centralized storage is problematic at this site, and excavations in several structures in sector 4, a complex of 501 small rectangular buildings, yielded evidence of variable domestic use.33 Although traditionally interpreted as a storage complex, the buildings in sector 4 might have served as temporary housing during periodic assemblies at the site in which labor tribute would be given by locals, feasts and ritual events performed, and exotic goods distributed by Wari elites. Even if these structures were intended for eventual use as storage facilities, the apparent lack of large-scale storage at Pikillacta suggests that any Wari attempt to convert the Cusco region into a centrally administered province was never realized. A lack of large-scale storage facilities at the site would contradict interpretations of centralized economic organization at the provincial level and would stand in marked contrast with Inca provincial administration practices. McEwan has described the Wari transformation of the Lucre Basin(4sorsipy peamen Jo umnasnpy ueotaUry ‘saDtAag Csesqr] Jo “Idaqy Asauzn0}a How the Incas Built Their Heartland from the seventh century onward.*4 In reconnaissance work, he identified Wari settlements at the sites of Minaspata, Qolque Haycuchina, and Waska Waskan, located a few kilometers to the south and southeast of Pikillacta. Surface pottery at these sites included Wari and local Wari Period styles, as well as waste flakes of turquoise and obsidian and bro- ken marine shell. McEwan believes that these sites were occupied by Wari settlers perhaps living among local populations, and he suggests that craft production was important in the habitational sites.>> Across the basin from Pikillacta, three small residential areas were identified close to the local shrine of Mama Qolla: Unca Puncu, Tukuwayku, and Morro de Arica. These sites show evidence of extensive local Wari Period pottery, with rare examples of the Wari style, leading McEwan to suggest that these were villages of local non-Wari populations.°° Finally, McEwan identifies four sites as controlling access into the Lucre Basin: Chokepukio, Muyurinapata, Rayallacta, and Mullimulliyog. Some of these sites include Wari-style walled enclosures, but McEwan notes a presence of local and imitation Wari pottery at Rayallacta, among others. Such sites appear to have included a habitational element and appear to be associated with monitoring access into the Lucre Basin from the Vilcanota Valley and Cusco Basin. A system of roads and cause- ways connects several strategic Wari sites with each other and with Pikil- lacta, signifying a centralized administrative infrastructure in the basin.37 Although McEwan’s reconnaissance of the Lucre Basin was not conducted following a systematic survey methodology, his observations certainly demonstrate that Wari colonization had a profound effect on the Lucre Basin, sustaining a complex settlement hierarchy and probably a large population of colonists and subordinate locals. Batan Urqu and the Huaro Area Archaeological research is just beginning in the Huaro area, located about 15 kilometers southeast of Pikillacta, but it is important to note some similarities to the Wari occupation of the Lucre Basin.* Excava- tions at Batan Urqu have yielded large quantities of Wari pottery, as well as tombs indicating residence by elite individuals with access to large quantities of goods from the Wari imperial heartland. Recent survey work by Mary Glowacki and colleagues has identified several habitation sites, some of them with Wari rectangular compounds.*° Preliminary research suggests that Batan Urqu may have been established before theWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 73 construction of Pikillacta was initiated, an occupation perhaps similar to that of Cerro Amaru in the Huamachuco region.*! Sites in the Huaro area have low percentages of Qotakalli pottery from the Cusco Basin, with considerable amounts of pottery from the Wari heartland, including the Okros, Chakipampa, Vifiaque, and Wamanga styles.42 Wari imperial pottery is rare in other parts of the Cusco region, and the architectural remains at Huaro sites suggest nonlocal settlement of the area by groups with regular contacts with the Wari heartland. A well- established Wari colony in the Cusco region would provide the means for developing relationships with some local groups so that labor tribute could be coordinated for the construction of Pikillacta and other major colonial infrastructure. Additional research is needed to clarify the nature of the Wari occupation of Huaro, but it is clear that the area between the Lucre Basin and modern Urcos was heavily colonized during the Wari Period and that some local populations were full partici- pants in the Wari colonial system. Wari settlers may have been active in craft production for exchange with local populations or with the imperial heartland (or for acquiring goods not available locally). It is also possible that, like colonists from other early states, Wari settlers initially moved into the Cusco region in search of agricultural lands to sustain themselves. Given the severe climatic fluctuations of the seventh century, some Wari “imperialism” might constitute a diaspora movement that capitalized on new agricultural techniques, in particular the development of irrigation canals and agricultural terraces for intensive maize agriculture.*? As the colonies grew and their administrative interests expanded, local populations probably provided labor to quarry stone and construct the architecture and road projects in the Lucre Basin. The segmented construction of walls at Pikillacta is consistent with rotational labor organized by local elites, possibly in rounds similar to modern mit’a (a system of rotating corvée labor).44 The Wari colonization of the Lucre Basin and Huaro area had an impact on the local landscape and its populations that many Wari researchers have interpreted as sustained and profound. The apparent lack of large-scale storage at Pikillacta raises the question of how directly local populations outside these areas were administered. If the Cusco region was an administered Wari province, the regional archaeology should indicate changes in political economy, revealed through settlement shifts and the intensification of agriculture. In addition, regional74 How the Incas Built Their Heartland settlement should have become more hierarchical through the establishment of secondary administrative sites. Wari material culture should be visible throughout the region, particularly in areas with strong signs of imperial infrastructure. None of these indicators has been observed in systematically surveyed areas in the Cusco region. The Scope of Wari Administration in Cusco The establishment of Wari colonies did not have a major impact on preexisting settlement patterns outside of the Lucre-Huaro area. The settlement system in the Chit'apampa Basin does not show major changes from AD 600 to AD 1000 (fig. 4.10). In the Sacred Valley study region, twenty of the forty-six sites with Qotakalli pottery have no Araway pottery, although most of these were hamlets that were abandoned over time. At least one small village was established during the Wari Period, and one appears to have been abandoned. Most villages with Qotakalli pottery also have Wari Period Araway pottery, indicating that the most noteworthy development in about four hundred years of Wari colonization in the Cusco region was a general nucleation of settlement in some areas. This is especially true for areas closest to Cusco, while sites in the Sacred Valley proper demonstrate little change or Wari influence. One site that highlights the lack of Wari administrative development in the Sacred Valley is Wanka, located approximately 15 kilometers from Pikillacta (fig. 4.11). The site is located within an easy three- to four- hour walk of Pikillacta, but it is probably too far from the Cusco Basin to have been administered by a prestate polity.45 Situated at an elevation of about 3100 meters, Wanka is ideally situated for the development of maize agriculture because of permanent water sources and a frost-free valley-bottom climate. If Wari imperialism in Cusco were based on direct territorial administration and the intensification of maize agriculture, one would expect Wari influence to extend down the Sacred Valley at least to Wanka. This influence would be demonstrated by the presence of Wari-style pottery and architecture at this site and others located near prime agricultural lands in the Sacred Valley (and outside of the region interacting regularly with the Cusco Basin polity). Surface collections from Wanka reveal an extensive Wari Period assemblage, but it is not characterized by a high percentage of Wari-style pottery. Surface pottery included Cusco styles (Qotakalli and Araway),WARI IMPERIALISM AND LOCAL POLITICAL COMPETITION _75, @:-— 4m © Under 1 hectare Under 1 hectare, with Araway ‘and other Wari Period style HB smat vitage (5h) Fig. 4.10. Araway/Wari Period settlement in the Sacred Valley study region local imitations of these styles, local Lucre Basin pottery, and several fragments of a style that may have been produced in the Paucartambo region to the north. Only a few Wari-style shards were collected at Wanka. The low frequency of Wari imperial pottery at Wanka is consistent with its distribution throughout the Sacred Valley study region. With the exception of a few isolated shards found at sites with large components of local Wari Period pottery, Wari-style pottery is almost completely absent. This is largely true for the Paruro region as well, where Wari-style pottery was found in very low percentages at eight of nine identifiedsurseg 9197] Aqavau oyp ut sistuojoo ueAy so Ajod uIseg o9snZ ay Jonuos ayy sapun_ Uuoaq aAey 01 10u suvadde aus ayp ‘KaqpeA pauses ay. ur spuey i 1a1E907| “IHS POLIDd L sequeyy “rf By,WARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 77 sites.4© The exception to this pattern is the site of Muyu Rogo, a small (2500-square-meter) site located close to modern Paruro. Bauer conducted surface collections and test excavations at this site, encountering hundreds of fragments of Wari-style pottery and camelid bones.47 The pottery included a high percentage of fancy bowls and drinking vessels, and one excavated bone yielded a calibrated date of AD 765-865, about two hundred years after the initial Wari occupation of the Lucre-Huaro area.48 ‘The average distance of Paruro Wari sites from the principal settlements of the Cusco Basin was over 30 kilometers, with six of nine sites lying in a cluster of settlements around Paruro. By comparison, Piki- Ilacta lies just 20 kilometers from Paruro, about a six- to eight-hour walk on well-established trails.4? Bauer's data suggest that the Wari presence in Paruro did not constitute formal political administration but rather an indirect influence, possibly limited to feasts associated with the ceremonial maintenance of a local shrine. The Wari appear to have exercised considerable influence within the Cusco Basin, although the evidence is lacking for political or economic reorganization of the local polity. The absence of major changes in settlement pattern and hierarchy contrasts with the intense Wari colonial presence in the Lucre-Huaro region, suggesting that formal provincial administration was never fully implemented. Within the Cusco Basin, Wari-style pottery from the Ayacucho region is rare and is never present as the dominant pottery style at any site. Instead, as has already been noted, a local imitation of Wari styles called Araway was produced in the Cusco Basin, using clay sources similar to those used in the production of Qotakalli pottery.5° Araway pottery is found at nearly ninety sites in the Cusco Basin, as well as forty-two sites in the Sacred Valley study region and eight sites in the Paruro survey region.5! Within the traditional sphere of Cusco political control, Araway pottery (and the occa- sional Wari vessel) appears to have been distributed through Cusco elites, who probably would have had direct ties to Wari colonists and may have helped to mobilize labor tribute at certain times of the year. The stability of settlement systems and exchange networks for decorated pottery suggests that Wari administrators did not transform local economic production or bypass existing elite hierarchies to govern at the local level. Instead, Cusco Basin elites adopted Wari design motifs and distributed their local imitation of Wari pottery (Araway) throughout the area under their control. Wari elites had access to goods (craft items78 How the Incas Built Their Heartland and metal) that were not available locally, and the Cusco polity probably came under Wari hegemony because local elites desired to obtain such goods and participate in Wari ceremonial life. The Effects of Wari Imperialism on Inca State Formation The Wari occupation of the Lucre Basin and Huaro area represents a major development of a peripheral region, but one of very limited geo- graphic scope, similar to the Wari occupation of sites around Cerro Baul.5? The archaeological evidence indicates that Wari colonies were established in areas where they could have contact with several other regional polities. These colonies transformed their immediate environments, developing a complex administrative infrastructure. Existing patterns of interregional exchanges for goods like obsidian were interrupted,53 and Wari officials appear to have co-opted ritual life in some places. Still, settlement patterns from the Cusco region indicate that Wari’s transformative effect was not felt locally in the Cusco Basin and areas under its influence. Local elites probably had close interactions with Wari colonists and may have helped to organize labor tribute for Wari projects. It appears that these elites used such interactions to their own advantage, redistributing Wari status goods throughout their political networks and even using established craft production systems to imitate Wari fancy pottery (fig. 4.12). Itis possible that the Wari polity attempted to formalize its hegemony as a means of developing more intensive tributary relationships with potential provinces. While the Wari collapse is poorly understood, the Cusco data may indicate local resistance to increased imperial administration. If so, local elites were undoubtedly involved in developing regional alternatives to the Wari system.5+ The decline of the Wari empire stimulated the processes leading up to the formation of the early Inca state. Centuries of elite interaction during the Wari occupation must have had an effect on how elite status was communicated, even though Wari administrative control remained limited at the local level. The Wari influence on local sources of power was probably felt in political, economic, and religious/ideological ways. The practice of providing labor tribute for the construction of state facilities (including terraces, canals, and roads) probably increased the amountWARI IMPERIALISM AND LOCAL POLITICAL COMPETITION 79 above 4000 mas! 4 (Coca and Lowland SS Products POLITY COLONIES Tribute? Interaction with Titicaca Basin ™ Wari-dominated village A Independent village EEE SE 10 20km Fig. 4.12. A schematic of Wari influence in the Cusco region, Wari’s control outside of the Lucre Basin and Urcos area appears to have been indirect. of labor tribute that local elites could extract from their supporters. Economically, the intensification of agricultural production, devotion to maize agriculture, and improvement of agricultural lands created a reliable surplus in maize not witnessed in earlier local polities. Participation in Wari state religious rituals introduced local elites to more hierarchical and cosmologically overarching religious systems and imbued them with more authority than they might otherwise have had. In short, Wari imperialism in Cusco had few local effects adminis- tratively, but far-reaching effects in the conceptualization of elite authority. The decline of the Wari was probably due in part to local elite80 How the Incas Built Their Heartland desertion of the existing tributary system. Future research should eluci- date whether this was because the Wari were in decline and local elites required a new expression of local authority or because of an active challenge of imperial authority by local elites, including Wari provin- cials. Either way, the abandonment of Wari’s colonies created a political vacuum in Cusco, with a large number of petty polities led by elite leaders anxious to dominate their rivals.CHAPTER FIVE Setting the Stage for Inca State Formation (AD 1000-1200) Wari colonies dominated the Cusco region for three hundred years or more, and their abandonment signals a major change in regional political organization. After AD rooo, settlement patterns and site hierarchies that had been relatively stable for centuries were transformed radically. In some areas, small valley-bottom villages were abandoned as new polities regrouped around defensible locations. For example, in the Sacred Valley study region, less than 20 percent (41/224) of the sites occupied between AD 1000 and 1400 (a period referred to as the Killke Period) have evidence of continuing settlement from the preceding period of Wari hegemony. Such changes are markedly different from the Cusco Basin, where most villages continued to be inhabited, growing larger as major Wari sites elsewhere in the region were abandoned. Settlement shifts in the Sacred Valley signal a new environment of political competition between several small complex polities, a situation culminating in the formation of the Inca state in the Cusco Basin. This chapter first considers the decline of the Wari and Tiwanaku empires, then looks at the transformation of settlement in the Cusco region, comparing settlement patterns in the Sacred Valley study region to those from the Cusco Basin, the Lucre Basin, and other parts of the region. The Decline of the Wari and Tiwanaku Empires Archaeological data from the Wari and Tiwanaku heartland and provincial regions reveal that both empires began to decline between about AD goo and AD 1000, although processes of disintegration were not uniform throughout the Andes.! There is evidence suggesting that climatic stress played a role in political collapse, but some researchers believe that internal social competition led to rebellions that interrupted political 8r82 How the Incas Built Their Heartland integration. In some provincial regions, imperial installations were abandoned at this time, and some of these were even burned.* Other areas saw the persistence of imperial colonies or enclaves, most of them now independent from administration by the capital.* Processes of balkanization apparently characterized much of the An- dean region from AD rooo to AD 1200.5 The Wari and Tiwanaku heartlands, incapable of integrating competing elite groups or funding centralized state governments, broke down into numerous petty polities. In the Titicaca Basin, several small Aymara polities (se/torios) emerged and competed for regional prominence.$ Settlement in the Titicaca Basin after about AD 1100 became less hierarchical, characterized by a shift to clusters of small, undifferentiated hamlets and villages scattered around large ridgetop settlements that were surrounded by concentric walls.” Decline of the Imperial Heartlands By around AD 1100, settlement patterns in the Tiwanaku Valley indicate a population decline at higher-order administrative sites, with a growth in small sites dispersed throughout all environmental zones— evidence for localization of production and decline of centralized administration.* Marc Bermann has identified a return to single-family residence patterns and the possible disuse of administrative or public architecture at this time.? The state capital was abandoned by AD 1200. Although extensive settlement data have not been published for the Ayacucho region, the Wari heartland appears to have undergone similar processes of decentralization at about the same time. The Wari capital may have been sacked around AD 1100, and the political unity of the Wari heartland gave way to the development of small petty polities organized around defensible settlements.'° Areas just outside the imperial heartland, where clear Wari settlement patterns and infrastructure have been identified, also saw dramatic settlement shifts from low-elevation or valley-bottom sites to higher-elevation or ridgetop settlements.!! Both in the Titicaca Basin and the highlands around Ayacucho, small polities were organized to meet corporate defensive needs, and many new settlements were established on ridges or hilltops that provided protection, visibility, and access to more environmental zones that would permit self-sufficient local economies. Even in the absence of a centralized state government, elites in some of these petty polities might have re- tained some aspects of preexisting ideologies, social stratification, andSETTING THE STAGE FOR INCA STATE FORMATION 83, the trappings of elite life—for example, political titles, the display of exotic goods, observation of sumptuary rules, and feasting.'7 Processes of Balkanization and Secession in the Wari and Tiwanaku Provinces Many areas with evidence of Wari settlement underwent political reorganization beginning in the tenth century.'? In the highlands, these are typified by radical shifts in settlement and a simpler political organization. On the coast, the assertion of local autonomy actually led to the development of new complex polities, and challenges to Wari administration appear to have led some provincial or local elites to break away from the heartland. The establishment or urban growth of sites like Chan Chan, Marca Huamachuco, Tiicume, Batan Grande, and possibly Cusco corresponds to the decline of Wari influence in these regions (AD 900-1100) and may represent cases where elites promoted local self-sufficiency as a viable administrative alternative to the Wari imperial system.'4 The Wari presence along parts of the north and central coast gave way to the formation of new states between AD rooo and AD 1200. The Sican polity represents one case of secondary state formation, where a new centralized polity came to control parts of the north coast from AD 900 to AD 1100, when most of the Andean region was experienc- ing political breakdown. In the Chincha Valley on the central coast, the decline of Wari influence led to a shift in settlement from control points along a valleywide irrigation system to locations closer to the coast. As the Chincha state developed, several large political and religious centers developed on the flood plain, linked by an intravalley road." Balkanized Polities of the Pacific Coastal Valleys The Pacific coastal valleys of southern Peru and northern Chile provide some of the most detailed archaeological data on local Wari and Ti- wanaku political balkanization after AD rooo. In the Moquegua Valley, the maize-producing elevations of the middle valley had been colonized by both Tiwanaku and Wari, and there appears to have been limited interaction between the outposts of the two empires until perhaps after AD 800.!7 Tiwanaku settlement was concentrated near the flood plain in the midvalley region, where large areas could be watered using simple canals. Tiwanaku influence in Moquegua appears to have stretched84 How the Incas Built Their Heartland almost to the Pacific coast, and a foreign presence has been identified in the Azapa Valley of northern Chile, where Tiwanaku colonists coexisted with local groups.'® The Wari occupation of Moquegua was concentrated around the sites of Cerro Bauil, Cerro Mejia, and Cerro Petroglifo, located above the midvalley Tiwanaku settlements in an area brought under cultivation through the construction of long irrigation canals and hillside terraces.'? ‘The collapse of the Tiwanaku colonial system in Moquegua appears to have occurred by about AD 950 and resulted in the abandonment and deliberate burning of large provincial settlements and a migration of some groups to the coast or upper valley.2° The Wari religious-administrative precinct at Cerro Bauil was burned and deserted a century or so later, although the excavators interpret this as an orderly abandonment.?! From AD 1000 to 1200 the coastal valleys of the south-central Andes saw the development of a number of distinct local ceramic styles, including Chiribaya, Tumilaca, San Miguel, and Pocoma. Many of these are associated with distinctive architecture, mortuary style, and textile patterns.?? It appears that several new ethnic and political groups developed at this time, some of them (like the Chiribaya and Tumilaca) probably including former Wari and Tiwanaku colonists, while others, like those living in the Azapa Valley of northern Chile, comprised indigenous populations.23 In the Moquegua Valley, Tumilaca sites continued to cluster around Cerro Baiil, but a series of new villages was established along the Tumi- laca River, where agriculture would be less dependent on long-distance irrigation systems.?+ Settlement patterns from coastal surveys indicate clusters of relatively small polities. Burials from the Chiribaya culture of the lower Moquegua Valley reveal that local elites had access to precious metals and other exotic goods.” It is clear that highland and lowland regions maintained exchange relationships after Wari and Tiwanaku declined but that these were not based on administrative control or tributary obligations. There was considerable movement of exotic goods along the coast, as well as between coastal valleys and the highlands. The Decline of Wari and Establishment of Local Polities in Cusco According to Gordon McEwan, the site of Pikillacta was abandoned sometime after AD 900, a process that he describes as rapid yet orderly.?°SETTING THE STAGE FOR INCA STATE FORMATION 85 The ongoing construction of the site was halted, and the occupants of Pikillacta sealed off some parts of the site as they systematically cleaned out rooms and subfloor offerings.2” Much of the site was subsequently burned.?* The sudden abandonment of Pikillacta followed about three hundred years of stable settlement and coincided with processes of decline and abandonment at other Wari sites.?2? Given that the Wari heartland was itself in decline, it is unlikely that entire populations from distant, long- standing colonies would have relocated to the capital. Instead, the population movements observed in the Cusco region indicate that it is more likely that Wari colonial populations and the local groups under their control fragmented and dispersed throughout the region, setting up independent polities that competed with each other for regional domination. The effects of the Wari collapse on the Lucre Basin have not been explored through systematic surveys, although researchers have conducted excavations at the post-Wari sites of Minaspata and Chokepukio (fig. 5.1).3° Not all Wari Period sites in the Lucre Basin were abandoned when Pikillacta was depopulated, but the scale of emigration from the basin is unknown because we still lack data on the size of Pikillacta’s permanent population. Settlement patterns to the north and east of the Cusco Basin indicate that some major settlement shifts occurred as Pikillacta was abandoned (fig. 5.2). The small valley-bottom agricultural villages in the Oropesa Basin (located between the Cusco and Lucre basins) were abandoned, and that area was almost completely depopulated from about AD 1000 to AD 1400. Fertile valley-bottom maize lands ceased to be farmed, and the only Killke Period settlement was Pukara Tipén, a large, nucleated village located several hundred meters above the valley floor in an area where the village and its lands could be surrounded by a massive defensive wall.>! A similar buffer zone also can be observed in the Sacred Valley study region, as the small villages and hamlets of the Chit’apampa Basin were abandoned in favor of settlement of larger, nucleated villages in more defensible locations. The Sacred Valley saw a substantial influx of population after AD 1000, with the establishment of numerous large villages on high ridges. Given the modest number and sizes of earlier settlements, it appears that the decline of the Wari colonies was associated with major demographic and political restructuring in the Sacred Valley. After AD 1000, the Sacred Valley was home to several small complex polities. Some maintained regular contacts with the Cusco Basin, while others(Arorsipy peavey Jo wmasnpy ueotIaUTy ‘S29TAI9g sear yo adaqq sar 1 Ast MO] E WO Parero] ‘orIndaxoyD -mnog ‘6rgh€E ‘ou aaneBayy) (punouSasof) eIETIP|EA WOH 85059 er ee & 2 eesSETTING THE STAGE FOR INCA STATE FORMATION 87 @cusco CUSCO BASIN. ny q SEN DEPOPULATED a BUFFER gi Tip! Chokepukio 3 Wit ‘Minaspata, YY LUCRE IS aN ot ae 2 q nut ZOAZC — as © 0 10 20km Fig. 5.2. Killke Period settlement in the Lucre Basin and Urcos area. Several large, nucleated villages or towns grew as the Wari colonies were abandoned around AD. 000. developed distinct local architectural, mortuary, and pottery styles. Such distinctions in local material culture are indicative of differences in political or ethnic identity. Other archaeological research from the Cusco region indicates the formation of several small polities, organized around nucleated, often defensible, settlements. For example, reconnaissance by Kenneth Heffernan and Ann Kendall has identified such polities form- ing to the west of Cusco.* Evidence for Wari administration in these areas has not been reported to date, although much of the region remains to be surveyed intensively. In contrast to the many small settlement systems clustered around defensively sited principal villages, most of the Cusco Basin after AD 1000 appears to have been integrated into a single settlement system in which site location was influenced more prominently by economic production88 How the Incas Built Their Heartland pt “SP e Pungurhwayila Minaspats Vie * based on Bauer and Covey 2002 # .25-1.00ha. 0 10 20km ™ 1.00-7.00 ha. @ over 10ha. @® @ over50ha. |. Area of Sites with Burnt Daub on the Surface Il, Urban Cusco and Its Satellite Villages Ill, Area of New Killke Period Terraces and Villages IV. Area of Remote Killke Period Villages Fig, 5.3- Killke Period settlement in the Cusco Basin. Survey work by Brian Bauer has demonstrated a hierarchical settlement system with a high population density. than defense.> Settlement patterns from the Cusco Basin indicate increasing settlement density after AD 1000 (fig. 5.3). The depopulation of intermediate areas may partly account for the increased number and size of villages in the basin, as groups living between rival polities cast their lot with one or the other. Political complexity continued in the Cusco Basin after AD 1000, with major increases in population that were probably concomitant with increased competition between the Cusco Basin polity and regional rivals. The abandonment of Pikillacta was part of a massive transformation of regional settlement resulting in the establishment of numerous com-SETTING THE STAGE FOR INCA STATE FORMATION 89 peting petty polities. Given the uncertainty regarding settlement at the site of Cusco prior to about AD 1000, it would be problematic to posit a direct link between the decline of Pikillacta and the rise of the early Inca polity in Cusco. Settlement continuities in the Lucre Basin argue against a complete relocation of Wari settlement into the Cusco Basin, and it would seem reasonable that the development of Cusco as a paramount center was a local initiative, perhaps a reaction to the attempted intensification of Wari administration in the region. The establishment of a depopulated buffer zone between the Cusco Basin and Lucre Basin would appear to indicate a lack of shared traditions between the two areas, rather than a large-scale migration of post-Wari settlers to Cusco. After AD 1000, the populations of the Cusco Basin and Sacred Valley grew substantially, and the site of Cusco itself increased in size and population density. Polity and Ethnicity in the Sacred Valley after AD 1000 fall areas surveyed intensively, the Sacred Valley study region provides the best systematic evidence for massive settlement shifts related to the Wari decline. The timing of this shift appears to be at the very end of the Wari Period, evidenced by the presence of a small component of Araway pottery at thirteen Killke Period sites where no Qotakalli pottery was present. Generally, there is little continuity between Cusco-dominated Wari Period styles and Killke Period settlements, of which 83.5 percent (183/224) have no occupation from AD 400 to AD 1000. As the dispersed agricultural villages located on the lower slopes and valley-bottom areas were abandoned, new nucleated villages were established on remote ridgetops and in locations with good visibility and natural defenses. Twenty-one villages of 3 hectares or larger were identified in the Sacred Valley study region, located at an average elevation of 3928 meters—almost 1000 meters above the floor of the Sacred Valley (fig. 5.4). This movement represents a major change in site catchment, implying a different political economy and new concerns for community defense. Over two hundred small villages and hamlets were found at lower elevations (3574 meters for sites under a hectare, 3589 meters for sites 1-3 hectares), suggesting that, in many cases, large villages located in suni or puna lands oversaw agricultural production at lower elevations. This pattern of ridgetop settlements does not appear in Brian Bauer’s90 How the Incas Built Their Heartland ) Reasfana ©—.. @ Under 1 hectare-not shown HM 1-5hectare villages 5-10 hectare villages @ over 10 hectares Muyuchurqu i Sh Fig. 5.4. Major Killke Period settlements of the Sacred Valley study region. The largest sites settled after AD 1000 are located at elevations of 4,000 meters or higher. Paruro survey, but there is some evidence of remote or defensible settlements at high elevation in the southern part of the Cusco Basin.34 Ic is difficult at present to subdivide the pottery in use between about AD 1000 and AD 1400. A style called Killke was produced in the Cusco Basin at this time, but the process of collecting chronometric dates from stratified excavation deposits has not advanced sufficiently to permit ar- chacologists to identify a clear assemblage for the period from AD tooo until the formation of the Inca state around AD 1200-1300. The appearance of Inca-s yle pottery and architecture is also poorly under-SETTING THE STAGE FOR INCA STATE FORMATION or stood chronologically and overlaps with Killke ceramic use in the fourteenth (and perhaps the fifteenth) century, making it difficult at this time to use pottery alone to establish calendar dates for the process of state formation. Despite these limitations, it is clear that this period marks a radical change in the social organization and political economy of different groups in the Cusco region. Rank-size graphs for the Sacred Valley study region indicate a more hierarchical settlement pattern than was seen for preceding periods (fig. 5.5). The rank-size curve for the Sacred Val- ley for the period from AD 400 to AD 1000 is convex, consistent with conditions of upper-tier competition and a lack of central coordination.>® From AD tooo to AD 1400, the rank-size graph indicates a more linear distribution when excluding the city of Cusco and its secondary administrative center, established at Pukara Pantillijlla around AD 1300. Ex- cluding Inca administrative sites, it appears that this region had a more complex and well-integrated settlement system. Political Units in the Sacred Valley Estimated territory sizes for the largest villages after AD tooo are consistent with an administrative complexity below the state level, and it is possible to delimit the territories of several possible polities based on clusters of small and large villages. Site locations during this period are consistent with the exploitation of diverse animal and plant resources. Large villages like Qhapaqkancha, Kaytumarka, Markasunay, and Muyuch’urqu were situated at 4000 meters or higher, within easy reach of pasture lands for herding and agricultural lands for the production of tubers and Andean grains (fig. 5.6). The largest villages in the Sacred Valley study region appear to have been the principal settlements of small complex polities. Local elites would have managed a mixed agricultural and herding economy and probably were involved in the organization of labor to construct irrigation canals—for example, the one bringing water to the site and fields of Kaytumarka—and defensive works, like the walls at Muyuch’urqu or the defensive ditch at Raqchi (fig. 5.7). Several of the largest settlements were located along important routes between kichwa valleys and the montafia, and local elites probably organized camelid caravans to obtain lowland agricultural products. These small polities show evidence of a concern for community100,000}- 4 a ‘AD 1000-1400 > 2 w 10,000 N a Ww E a ™ 1,000; | 100 L H 1 10 100 RANK 1,000,000 |, ., AD 1400-1550 E 100,000 el s 2 ‘AD 1000-1400 Ww N a E 10,000 +- a 1,000 1 10 100 RANK Fig, 5.5. Rank-size graphs of the Sacred Valley study regionSETTING THE STAGE FOR INCA STATE FORMATION 93 0 70m Fig, 5.6, Plan of Markasunay, a large Killke Period village with early Inca architecture protection against raiding and warfare. Not only were many large villages situated in inaccessible locations, but some of these also had walls or defensive ditches, indicating that community or polity labor was being invested in defense. In addition to inaccessible or defensive settlement locations, there are other possible indicators of conflict between groups. Burnt wall daub, evidence of conflagrations that might have been associated with raiding, is almost exclusively associated with Kil- Ike Period villages outside of the Cusco sphere of influence. Burnt daub is absent in the large valley-bottom villages of the Cusco Basin but was found at several sites around Poroy and Cachimayo to the west of Cusco. In the Sacred Valley study region, burnt daub was observed on the surface at twenty-eight sites, all of which had substantial components ofeseyng 1e uownpi9s Jo uorsuedxs vouy a¥p 940}9q UDMID|NAS PoLIDd D>{[EY 98L] 942 “nban,yonsnyy re s|Jem DAISUD}Oq]SETTING THE STAGE FOR INCA STATE FORMATION 95 @ Burnt Daub Observed Fig. 5.8. Map of the location of burnt daub, possibly indicating the burning of settlements in warfare or raids after AD 1000 Killke Period pottery (fig. 5.8). A midden at Pukara Pantillijlla that dated to the thirteenth century contained sling balls and two stone axes, which might also serve as indirect evidence for conflict. Frequent raiding and warfare are common themes in many of the Inca chronicles for the prestate period, as discussed in greater detail in chapter 6. A system of balkanized, warring petty polities has been described as the prevailing condition for the Andean highlands prior to the Inca con quest.37 The chronicler Felipe Guaman Poma de Ayala conceptualizes the pre-Inca period in terms of decentralization, perpetual warfare, and irregular participation in regional ethnic confederations. The area96 How the Incas Built Their Heartland surrounding the developing Inca state contained multiple political groups of varying organizational complexity, and the Incas would take advantage of ethnic and political variability to expand and consolidate their imperial heartland. Ethnic Variation in the Cusco Region In some cases “ethnic” distinctions were not based on differences in material culture or language, and there is reason to believe that many— but not all—of the small groups living around Cusco were Quechua speakers.38 In contrast to their neighbors, Lucre Basin groups like the Muyna and Pinagua may have been more closely related to their Wari predecessors than to indigenous Cusco Basin groups. Variations in mortuary treatment and architectural style in the Sacred Valley allow us to distinguish between some populations. In the upper side valleys to the north of the Vilcanota River, some local groups placed burials in small ovoidal mortuary structures, suggesting either an influx of nonlocal groups or the adoption of a new mortuary style by some groups in the region (fig. 5.9).3° Groups in the Chit’apampa Basin, Pisaq area, and San Salvador area did not use such structures but buried their dead in simple cliff tombs. Cliff tombs below the site of Kaytumarka took the form of painted adobe boxes, while different styles of cliff burials were observed both in the Pisaq area and around Ancasmarca. Mortuary patterns indicate that groups living to the south of the Vilcanota River probably shared some cultural practices not observed among groups living to the north of the valley. The diversity of burial practices reveals that considerable cultural variation could be found within a very small region (fig. 5.10). Similar variations can be seen in the domestic architecture and pottery of the Cusco region. Distinct pottery styles have been identified for the Chongo Basin, the Lucre Basin, the Chinchero area, and the area around Colcha, and itis likely that other subregions also produced their own pottery. In areas where Killke Period architecture is well preserved, it is possible to observe variations in domestic architecture. Some sites are characterized by rectangular structures, others by ovoid or semi- circular forms. Distinctive construction features like corbeled doorways, multistory construction, and interior wall niches have been identified at the site level.4! In the Sacred Valley study region, groups living to the south of the98 How the Incas Built Their Heartland : @:—. him TD surtace Bone GAA Aboveground Structure A. cit structure © Git Tomb Fig. 5.10. Mortuary sites identified in the Sacred Valley study region Vilcanota River appear to have been more closely related to the Incas, while some of the populations living to the north of the river appear to constitute one or more distinct ethnic groups. The prevalence of Killke Period pottery from the Cusco Basin is markedly different on the north and south sides of the Vilcanota River. Of the twelve wares identified from Killke Period excavations at Pukara Pantillijlla, four (designated numbers 9-12) are thought to have been produced in the Cusco Basin, based on similarities in paste, temper, and wall thickness. These wares made up the majority of the decorated pottery collected at sites in the Chit’apampa Basin and along the south rim of the Sacred Valley (78 and 66 percent, respectively). To the north of the Vilcanota Valley, these waresSETTING THE STAGE FOR INCA STATE FORMATION 99 constituted far less of the decorated pottery collections, as low as 33 percent in the Qochog Valley above Calca and 41 percent in the Chongo Basin. Along with differences in mortuary treatment and domestic architecture, it appears that the Vilcanota Valley represented a major cultural divide prior to Inca state formation. Intensification and Diversity in Pre-Inca Political Economies The establishment of a number of small polities after AD 1000 involved substantial changes in Cusco’s regional political economies. From AD 600 to AD ooo the Wari colonies had developed intensive maize agriculture in the Lucre Basin and Huaro areas and included some locals in the use of maize—both as a food and fermented alcoholic beverage—in feasting and ritual events of a scale beyond what local elites could arrange. The location and distribution of Wari settlement and cultural remains suggest a production system based on the intensive local exploitation of maize, supplemented by other crops. Herding may have been managed through Pikillacta, given the presence of large rectangular enclosures with no interior architecture that McEwan interprets as corrals.4? Small storage facilities identified at Pikillacta would have held surpluses that could be mobilized to fete laborers and buffer Wari colonists against bad years, although province-level storage does not appear to have been in use at the site. The absence of evidence for large-scale production and storage of agricultural surpluses indicates that Pikillacta did not centrally manage provincial agricultural production and was not responsible for addressing resource fluctuations at the local level. It appears that Wari settlers relied on intensive maize production on improved lands but that they did not amplify production levels to the point that storage would be the primary risk-reduction strategy for the colonies and nearby local populations. It is possible that exchanges of exotic and craft goods would have provided another means for obtaining food and that the maintenance of caravans would have been an alternative to some staple production. This would require increased levels of agricultural production by local groups, who would then exchange agricultural surpluses for craft goods or exotic goods. While populations in the Cusco Basin and Lucre Basin lived in close proximity to resources for intensive maize agriculture, a substantial component of the Cusco region’s population met the challenges of a highly

COVEY, A. How The Incas Built Their Heartland. State Formation and The Innovation of Imperial Strategies

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

COVEY, A. How The Incas Built Their Heartland. State Formation and The Innovation of Imperial Strategies

Enviado por

Direitos autorais:

Formatos disponíveis

Você também pode gostar